Portal de Periódicos da CAPES

A population-based study measuring the prevalence of chronic kidney disease among adults in West Malaysia

2013; Elsevier BV; Volume: 84; Issue: 5 Linguagem: Inglês

10.1038/ki.2013.220

1523-1755

Lai Seong Hooi, Loke Meng Ong, Ghazali Ahmad, Sunita Bavanandan, Noor Ani Ahmad, Balkish Mahadir Naidu, Wan Nazaimoon Wan Mohamud, Muhammad Fadhli Mohd Yusoff,

Blood Pressure and Hypertension Studies

In this population-based study, we determine the prevalence of chronic kidney disease in West Malaysia in order to have accurate information for health-care planning. A sample of 876 individuals, representative of 15,147 respondents from the National Health and Morbidity Survey 2011, of the noninstitutionalized adult population (over 18 years old) in West Malaysia was studied. We measured the estimated glomerular filtration rate (eGFR) (CKD-EPI equation); albuminuria and stages of chronic kidney disease were derived from calibrated serum creatinine, age, gender and early morning urine albumin creatinine ratio. The prevalence of chronic kidney disease in this group was 9.07%. An estimated 4.16% had stage 1 chronic kidney disease (eGFR >90ml/min per 1.73m2 and persistent albuminuria), 2.05% had stage 2 (eGFR 60–89ml/min per 1.73m2 and persistent albuminuria), 2.26% had stage 3 (eGFR 30–59ml/min per 1.73m2), 0.24% had stage 4 (eGFR 15–29ml/min per 1.73m2), and 0.36% had stage 5 chronic kidney disease (eGFR 90ml/min per 1.73m2 and persistent albuminuria), 2.05% had stage 2 (eGFR 60–89ml/min per 1.73m2 and persistent albuminuria), 2.26% had stage 3 (eGFR 30–59ml/min per 1.73m2), 0.24% had stage 4 (eGFR 15–29ml/min per 1.73m2), and 0.36% had stage 5 chronic kidney disease (eGFR <15ml/min per 1.73m2). Only 4% of respondents with chronic kidney disease were aware of their diagnosis. Risk factors included increased age, diabetes, and hypertension. Thus, chronic kidney disease in West Malaysia is common and, therefore, warrants early detection and treatment in order to potentially improve outcome. The prevalence of chronic kidney disease (CKD) in Malaysia is unknown.1.Hooi L.S. Wong H.S. Morad Z. Prevention of renal failure – the Malaysian experience.Kidney Int. 2005; 67: S70-S74Abstract Full Text Full Text PDF PubMed Google Scholar There has been an increasing trend in dialysis provision for end-stage renal disease in Malaysia from 96 per million population in 2002 to 182 per million population in 2011.2.Lim Y.N. Ong L.M. Goh B.L. 19th Report of the Malaysian Dialysis and Transplant Registry 20112012www.msn.org.myGoogle Scholar By the end of 2011, there were 27,572 patients on renal replacement therapy (RRT) in Malaysia (prevalence rate of 966 per million population). RRT places a large burden on the health-care budget.3.Couser W.G. Remuzzi G. Mendis S. et al.The contribution of chronic kidney disease to the global burden of major communicable diseases.Kidney Int. 2011; 80: 1258-1270Abstract Full Text Full Text PDF PubMed Scopus (913) Google Scholar It is important to obtain accurate local data on CKD to facilitate health-care planning including review of health-care priorities, program activities, and allocation of resources. The objective of this population-based study is to determine the prevalence of CKD in adults aged ≥18 years in West Malaysia. There were 876 respondents who agreed to participate out of the 1152 targeted, giving a response rate of 76%. Table 1 compares the overall National Health and Morbidity Survey (NHMS) 2011 cohort in West Malaysia aged ≥18 years (N=15,147) with the CKD substudy cohort (N=876). There was no difference in the profile of both cohorts.Table 1Comparison between National Health and Morbidity Survey (NHMS) cohort sample and CKD sampleNHMS cohort (≥18 years), N=15,147CKD sub-study, N=876Sociodemographic characteristicsNumberMeanMedianNumberMeanMedianIQRAge (years)15,14742.241.087642.943.022.0Systolic blood pressure (mmHg)14,631129.5127.0868130.2127.523.5Diastolic blood pressure (mmHg)14,63080.280.086880.980.016.5Glucose level (mmol/l)13,4366.15.47935.775.41.4Cholesterol level (mmol/l)13,7424.94.88284.04.81.4Abbreviations: CKD, chronic kidney disease; IQR, interquartile range. Open table in a new tab Abbreviations: CKD, chronic kidney disease; IQR, interquartile range. Table 2 shows the baseline characteristics of the CKD substudy population (N=876). Almost half of respondents (41.8%) were middle-income earners and 49.1% had a maximum education at the secondary school level. Microalbuminuria was detected in 0.5% of respondents with estimated glomerular filtration rate (eGFR) <60ml/min per 1.73m2 and in 9.1% of respondents with eGFR ≥60ml/min per 1.73m2 (total 9.6%). Macroalbuminuria was detected in 0.8% of respondents with eGFR <60ml/min per 1.73m2 and in 2.8% of respondents with eGFR ≥60ml/min per 1.73m2 (total 3.6%). There were 172 diabetics in the CKD substudy (that is, 19.6% of the cohort) and 32% of them had albuminuria. This was much higher than the prevalence of albuminuria in the entire CKD cohort (13.2%). Hypercholesterolemia was present in 44% and hypertension was present in 38.4%.Table 2Baseline characteristics of the CKD study population (N=876)Sociodemographic characteristicsN%Gender Male42047.9 Female45652.1Race Malay47053.7 Chinese18521.1 Indian17720.2 Others445.0Education attainment No formal education465.3 Primary17920.4 Secondary43049.1 Tertiary21624.6 Unclassified50.6StrataaUrban: population ≥10,000 per enumeration block (EB); rural: population 1000–10,000 per EB. Urban79190.3 Rural859.7Income quartilesbIn June 2011, 1 USD=3.0 RM (Malaysian currency ringgit). Low (<RM 2300 per month)31335.7 Middle (RM 2300–5600 per month)36641.8 High (≥RM 5600 per month)19722.5History of illness Previously diagnosed CKD80.9 Hypertension33638.4 Hypercholesterolemia38144.0 Diabetes17219.6 History of cardiovascular disease273.1 History of stroke80.9 Positive family history of CKD819.3NSAID use Yes16518.8 No71181.2 At least once a day526.0 At least once a week252.8 At least once a month384.3 Less than once a month465.3 No answer40.4Traditional medicine Yes12314.0 No75386.0 At least once a day566.4 At least once a week293.3 At least once a month313.5 No answer70.8Heart disease Yes273.1 No84996.9Family history of kidney disease Yes819.2 No78789.8Cigarette smoking Ever smoker24729.0 Current smoker16919.8 Noncurrent smoker789.2 Never smoker60571.0Alcohol consumption Ever drinker15918.2 Current drinker9510.9 Noncurrent drinker647.3 Never drinker71481.8Albuminuria Normoalbuminuria74386.8 Microalbuminuria (eGFR <60)40.5 Microalbuminuria (eGFR ≥60)789.1 Macroalbuminuria (eGFR <60)70.8 Macroalbuminuria (eGFR ≥60)242.8Albuminuria among diabetics Normoalbuminuria11768.0 Microalbuminuria3822.1 Macroalbuminuria179.9Physical activity Active53661.8 Inactive33238.2Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; NSAID, nonsteroidal anti-inflammatory drug.a Urban: population ≥10,000 per enumeration block (EB); rural: population 1000–10,000 per EB.b In June 2011, 1 USD=3.0 RM (Malaysian currency ringgit). Open table in a new tab Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; NSAID, nonsteroidal anti-inflammatory drug. In this cohort, 18.8% admitted to using nonsteroidal anti-inflammatory drugs, and 14% admitted to using traditional medicines. There were 29% of respondents who had ever smoked and 18.2% who had ever taken alcohol. Out of the 876 respondents, 850 had eGFR measurements (Table 3). In all, 78% of the respondents had eGFR ≥90ml/min per 1.73m2, 19.1% had eGFR 60–89ml/min per 1.73m2, 2.2% had eGFR 30–59ml/min per 1.73m2, 0.2% had eGFR 15–29ml/min per 1.73m2, and 0.4% had eGFR <15ml/min per 1.73m2.Table 3Prevalence of eGFR category (N=850)eGFR (ml/min per 1.73m2)NPrevalence (%)Normal (≥90)66478.1Mild (60–89)16219.1Moderate (30–59)192.2Severe (15–29)20.2Very severe (<15)30.4Abbreviation: eGFR, estimated glomerular filtration rate in ml/min per 1.73m2 using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Open table in a new tab Abbreviation: eGFR, estimated glomerular filtration rate in ml/min per 1.73m2 using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. At baseline, there were 113 respondents with albuminuria. However, only 59 (52.2%) agreed to be retested. Of the 42 subjects with microalbuminuria, 19 (45.2%) had a positive confirmatory test. Of those with macroalbuminuria, 70.6% (12/17) had persistent albuminuria on repeated testing (Table 4).Table 4Persistence of microalbuminuria and macroalbuminuriaNormoalbuminuria in the second specimenMicroalbuminuria in the second specimenMacroalbuminuria in the second specimenTotal, N=59Number%Number%Number%Microalbuminuria in the first specimen422354.81945.200Macroalbuminuria in the first specimen17529.4317.6952.9 Open table in a new tab Out of the 876 respondents, 840 had both eGFR and urine albumin creatinine ratio (ACR) measurements (Table 5). For CKD stages 1 and 2, the prevalence was adjusted according to the persistence of albuminuria. The distribution of CKD stages were as follows: stage 1, 4.16%; stage 2, 2.05%; stage 3, 2.26%; stage 4, 0.24%; and stage 5, 0.36% (total CKD prevalence is 9.07%). If persistence of albuminuria was not taken into account, the prevalence of CKD would be 15%.Table 5Prevalence of CKD stages (N=840)Albuminuria within each level, N (%)Persistence of albuminuria (%)Albuminuria within each level (%) (corrected for persistence)Estimated GFR (ml/min per 1.73m2)NoneMicroMacroMicroMacroMicroMacro≥90586 (69.8)57 (6.8)13 (1.5)45.270.63.071.0960–89128 (15.2)21 (2.5)11 (1.3)45.270.61.130.9230–59NANANANANANANA15–29NANANANANANANA<15NANANANANANANANumber of respondentsPrevalence of CKD (%)Adjusted prevalence of CKD (%)Normal714Stage 1708.334.16Stage 2323.812.05Stage 3192.262.26Stage 420.240.24Stage 530.360.36Total84015%9.07%Abbreviations: CKD, chronic kidney disease; GFR, glomerular filtration rate; NA, not applicable. Open table in a new tab Abbreviations: CKD, chronic kidney disease; GFR, glomerular filtration rate; NA, not applicable. CKD was found in a total of 126 of the respondents. Only 5 of them (4%) were aware that they had CKD, with 96% being unaware. Using univariate analysis, the factors associated with significantly increased risk for CKD were age, diabetes, hypertension, and hypercholesterolemia (Table 6). Logistic regression showed that the prevalence of CKD increased with age. Respondents with diabetes were 2.6 times more likely than nondiabetics to have CKD, whereas those with hypertension were 3.1 times more likely than nonhypertensives to have CKD (Table 7).Table 6Factors associated with CKD prevalence by univariate analysis (N=126)VariableN% With CKDP-valueUnadjusted OR (95% CI)Age 18–39226.2<0.0011 40–647718.83.50 (2.13–5.76) 65+2736.08.51 (4.49–16.13)Gender Male5112.60.0661 Female7517.31.45 (0.99–2.14)Race Malay7115.60.7981 Chinese2716.11.04 (0.64–1.69) Indian2212.70.79 (4.73–1.32) Others614.00.88 (0.36–2.16)Household income Low (<RM 2300 per month)4816.20.1800.73 (0.47–1.14) Middle (RM 2300–5600 per month)4412.41.13 (0.70–1.83) High (≥RM 5600 per month)3417.91Cigarette smoking Ever3012.60.1760.74 (0.47–1.15) No9416.31Alcohol consumption Ever2015.50.5320.83 (0.50–1.39) No10613.21Physical activity Active6913.50.0631 Inactive5717.71.38(0.93–2.02)Diabetes Yes5533.1<0.0014.16 (2.7–6.27) No6810.81Hypertension Yes9027.6<0.0015.05 (3.33–7.67) No367.01Hypercholesterolemia Yes7720.8<0.0012.37 (1.59–3.51) No46101Heart disease Yes415.40.5641.03 (0.35–3.04) No12215.01Family history of kidney disease Yes1419.20.1721.42 (0.77–2.63) No10914.31NSAID use Yes1811.10.7500.66 (0.39–1.12) No10815.91Traditional medicine use Yes1310.90.2130.66 (0.36–1.21) No11315.71Abbreviations: CI, confidence interval; CKD, chronic kidney disease; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; RM, Malaysian currency ringgit.Definition of chronic kidney disease status: estimated glomerular filtration rate (eGFR) <60ml/min per 1.73m2 and/or albuminuria. Open table in a new tab Table 7Adjusted OR for factors associated with CKDVariableAdjusted OR (95% CI)P-valueAge (years) 40–641.82 (1.04–3.19)0.036 65+3.15 (1.51–6.55)0.020Diabetes2.56 (1.63–4.01)P<0.001Hypertension3.09 (1.92–4.97)P<0.001Hypercholesterolemia1.29 (0.830–2.00)0.259Abbreviations: CI, confidence interval; CKD, chronic kidney disease; OR, odds ratio. Open table in a new tab Abbreviations: CI, confidence interval; CKD, chronic kidney disease; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; RM, Malaysian currency ringgit. Definition of chronic kidney disease status: estimated glomerular filtration rate (eGFR) 18 years old were on RRT in West Malaysia (Malaysian Dialysis and Transplant Registry, personal communication), which amounts to 0.17% of the population (total population at risk was 14,595,893). The estimate of 0.36% for stage 5 CKD is higher than the number of patients on RRT in the country at the time. Not all patients with eGFR <15ml/min per 1.73m2 were on RRT. The National Health and Morbidity Survey 2011 is a population-based cross-sectional study with stratified random sampling. The CKD substudy relied on experienced epidemiologists, trained field teams, and statisticians from the Institute for Public Health, Ministry of Health, Malaysia. There was a quality assurance program for both data collection and the central laboratory. The study took place over 4 months in the 11 states of West Malaysia and the Federal Territories of Kuala Lumpur and Putrajaya. Urine samples and serum creatinine were sent by courier to the central laboratory in Kuala Lumpur to arrive on the same day of sampling, which made it difficult to include the remote areas of West Malaysia and East Malaysia (the states of Sabah and Sarawak). The baseline characteristics of the CKD substudy were similar to that of the overall NHMS 2011 cohort (some data not shown). There was no difference in the profile of both cohorts, except for a preponderance of urban respondents in the CKD sample, which occurred by design (90.3% vs. 77.5%). This suggests that the CKD substudy was conducted on a valid representative sample for the population of West Malaysia. There was a good response rate of 76%.4.McCullough K. Sharma P. Tariq A. et al.Measuring the population burden of chronic kidney disease: a systematic literature review of the estimated prevalence of impaired kidney function.Nephrol Dial Transplant. 2012; 27: 1812-1821Crossref PubMed Scopus (91) Google Scholar The instruments used to determine blood sugar, cholesterol, and blood pressure had been standardized, validated, calibrated, and used throughout the study. The measurement of creatinine used a calibration traceable to the isotope dilution mass spectroscopy reference method. Care was taken to locate the subjects who worked in the daytime. The field staff worked both in the evenings and on weekends to locate the respondents. Hence, the number of men and women was almost equal. All the three main ethnic groups in Malaysia were represented in proportion to the general population. eGFR was derived from the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation5.Levey A. Stevens L. Schmid C. for the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) et al.A new equation to estimate glomerular filtration rate.Ann Intern Med. 2009; 150: 604-612Crossref PubMed Scopus (15959) Google Scholar using serum creatinine, age, and sex as variables. The CKD-EPI equation had been validated by Teo et al.6.Teo B.W. Xu H. Wang D.H. et al.GFR estimating equations in a multiethnic Asian population.Am J Kidney Dis. 2011; 58: 56-63Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar on a multiethnic population in Singapore. We chose to use the early-morning spot urine ACR, although it is a more costly and cumbersome method of estimating albuminuria, as it provides a sensitive measure for detection of kidney damage. The definition of albuminuria followed that stated in the KDIGO (Kidney Disease Improving Global Outcomes) guideline for CKD.7.Improving Global Outcomes (KDIGO) CKD Work Group KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease.Kidney Intl Suppl. 2013; 3: 1-150Abstract Full Text Full Text PDF Scopus (1578) Google Scholar It was not possible to repeat the measurement of albuminuria in every case, but 52.2% of initial positive results were retested. Microalbuminuria persisted in 45.2% and macroalbuminuria in 70.6%. For albuminuria, the rate of persistence was taken into account in the calculation of CKD stages 1 and 2. If persistence of albuminuria was ignored, the prevalence of CKD would have been higher (15 instead of 9.07%). The other signs of kidney damage, for example, structural changes on ultrasound and persistent hematuria, were not recorded, and the prevalence of CKD may be higher than estimated in the study. Serum creatinine was not repeated after 3 months, as this was not possible in a short-term survey. No tests for hematuria were instituted, as positive results would have had to be confirmed with microscopy, which is logistically difficult.8.Chadban S. Briganti E. Kerr P. et al.Prevalence of kidney damage in Australian adults: The Ausdiab Kidney Study.J Am Soc Nephrol. 2003; 14: S131-S138Crossref PubMed Google Scholar The prevalence of CKD in West Malaysia of 9.07% is similar to the rates in the region. The prevalence of CKD in Asia varies from 6.8% in South Korea, 12% in Taiwan, 13% in Beijing, and 13% in Japan to 17.5% in Thailand.9.Li P.K.T. Chow K.M. Matsuo S. et al.Asian chronic kidney disease (CKD) best practice recommendations: positional statements for early detection of CKD from Asian Forum for CKD Initiatives (AFCKDI).Nephrology (Carlton). 2011; 16: 633-641PubMed Google Scholar However, the prevalence depends on the definition of CKD, study design, and methodology.4.McCullough K. Sharma P. Tariq A. et al.Measuring the population burden of chronic kidney disease: a systematic literature review of the estimated prevalence of impaired kidney function.Nephrol Dial Transplant. 2012; 27: 1812-1821Crossref PubMed Scopus (91) Google Scholar In South Korea (for 5136 adults), the MDRD (Modification of Diet in Renal Disease) equation alone with eGFR <60ml/min per 1.73m2 was used to define CKD.10.Jang S.Y. Kim I.H. Ju E.Y. et al.Chronic kidney disease and metabolic syndrome in a general Korean population: the Third Korea National Health and Nutrition Examination Survey (KNHANESIII) Study.J Public Health. 2010; 32: 538-546Crossref Scopus (30) Google Scholar In Taiwan (for 462,293 adults), the MDRD equation was used to estimate GFR; for stage 1 and stage 2 CKD, proteinuria from dipstick adjusted for persistence was included.11.Wen C.P. Cheng T.Y. Tsai M.K. et al.All-cause mortality attributable to chronic kidney disease: a prospective cohort based on 462 293 adults in Taiwan.Lancet. 2008; 371: 2173-2182Abstract Full Text Full Text PDF PubMed Scopus (730) Google Scholar In Beijing, a large sample of 13,925 adults was tested, with eGFR measured using calibrated serum creatinine and a formula specific to China. Albuminuria, which was measured by albumin creatinine ratio, and hematuria, which was confirmed by microscopy, were included in the definition, based on single measurements.12.Zhang L. Zhang P. Wang F. et al.Prevalence and factors associated with CKD: a population study from Beijing.Am J Kidney Dis. 2008; 51: 373-384Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar The study from Japan in 504,724 adults used a Japanese equation for eGFR and dipstick for proteinuria to define CKD.13.Imai E. Horio M. Watanabe T. et al.Prevalence of chronic kidney disease in the Japanese general population.Clin Exp Nephrol. 2009; 13: 621-630Crossref PubMed Scopus (382) Google Scholar The study from Thailand in 3459 subjects used the MDRD equation with serum creatinine standardized with isotope dilution mass spectroscopy; for stage 1 and stage 2 CKD, the presence of a single estimation of albuminuria quantified by urine ACR and/or hematuria was included.14.Ingsathit A. Thakkinstian A. Chaiprasert A. et al.Prevalence and risk factors of chronic kidney disease in the Thai adult population: Thai SEEK study.Nephrol Dial Transplant. 2010; 25: 1567-1575Crossref PubMed Scopus (124) Google Scholar Only 4% of respondents were aware of the diagnosis. Awareness of CKD was found in 1.9% in Thailand14.Ingsathit A. Thakkinstian A. Chaiprasert A. et al.Prevalence and risk factors of chronic kidney disease in the Thai adult population: Thai SEEK study.Nephrol Dial Transplant. 2010; 25: 1567-1575Crossref PubMed Scopus (124) Google Scholar and 3.54% in Taiwan.11.Wen C.P. Cheng T.Y. Tsai M.K. et al.All-cause mortality attributable to chronic kidney disease: a prospective cohort based on 462 293 adults in Taiwan.Lancet. 2008; 371: 2173-2182Abstract Full Text Full Text PDF PubMed Scopus (730) Google Scholar This gives weight to the recommendation that screening and detection should be carried out in high-risk groups, especially in the primary-care setting. More public education about CKD is also necessary. The incidence of albuminuria in respondents with diabetes was 2.4 times higher than the substudy population (32% vs. 13.2%). As the overall prevalence of diabetes was very high (19.6%), this is an ominous finding, and interventions, such as glycemic control and angiotensin-converting enzyme inhibitors should be instituted to prevent progression of diabetic nephropathy. Local practice guidelines recommend yearly testing for albuminuria.15.Management of chronic kidney disease in adults. Clinical practice guidelines. Ministry of Health Malaysia 2011. MOH/P/PAK/217.11(GU). www.msn.org.my, accessed on 11 December 2012Google Scholar Medical personnel should adhere closely to this guideline. The percentage of new dialysis patients with diabetes mellitus had risen from 44% in 2000 to 56% in 2011.2.Lim Y.N. Ong L.M. Goh B.L. 19th Report of the Malaysian Dialysis and Transplant Registry 20112012www.msn.org.myGoogle Scholar The factors associated with CKD in other studies include age, diabetes, hypertension, metabolic syndrome, use of traditional medicines, and nonsteroidal anti-inflammatory drugs.9.Li P.K.T. Chow K.M. Matsuo S. et al.Asian chronic kidney disease (CKD) best practice recommendations: positional statements for early detection of CKD from Asian Forum for CKD Initiatives (AFCKDI).Nephrology (Carlton). 2011; 16: 633-641PubMed Google Scholar In this study, increasing age, diabetes, and hypertension were identified as significant risk factors. These factors would be considered ‘high risk’ in Malaysia, and early detection in these groups may ameliorate the progression of CKD, the associated complications, cardiovascular morbidity, and mortality.7.Improving Global Outcomes (KDIGO) CKD Work Group KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease.Kidney Intl Suppl. 2013; 3: 1-150Abstract Full Text Full Text PDF Scopus (1578) Google Scholar,16.Levey A. Coresh J. Chronic kidney disease.Lancet. 2012; 379: 165-180Abstract Full Text Full Text PDF PubMed Scopus (1293) Google Scholar This study was a substudy of the NHMS 2011,17.Institute of Public Health (IPH) 2011 National Health and Morbidity Survey 2011 (NHMS 2011), NMRR 10–757–6837. Volume I: Methodology and General Findings, 258 pages. ISBN 978 967 3887 67 5.Google Scholar which was carried out by the Institute for Public Health, Malaysia, to describe the health status, health-related behaviors, and health service utilization of a representative sample of the population of Malaysia. The study was conducted from April to July 2011. This study had received the approval of the Malaysian Research and Ethics Committee and was registered with the National Medical Research Register (NMRR 10-757-6837). An up-to-date and representative sampling frame for this population has been provided by the Department of Statistics. Malaysia is divided into enumeration blocks, that is, geographically continuous areas with identifiable boundaries. There were ∼75,000 enumeration blocks EBs in Malaysia in the year 2010. The sampling frame was stratified by state and urban/rural residence For the NHMS, a stratified two-stage cluster sampling design was used to draw a sample of 9258 private dwellings. For the CKD substudy, only respondents from West Malaysia were selected (Figure 1). Remote areas categorized as strata 4 by the Department of Statistics with a population <1000 per locality were excluded because of logistic restrictions. The sample size was calculated using a formula for a complex design study estimating population prevalence.17.Institute of Public Health (IPH) 2011 National Health and Morbidity Survey 2011 (NHMS 2011), NMRR 10–757–6837. Volume I: Methodology and General Findings, 258 pages. ISBN 978 967 3887 67 5.Google Scholar It was determined based on the expected prevalence of CKD of 10%, 0.03 margin of error, and 95% confidence interval, resulting in an optimal sample size of 770 for estimating prevalence at the national level. The final sample size was increased to 1152, taking nonrespondents into consideration. The inclusion criteria were all noninstitutionalized individuals above the age of 18 years who gave consent for participation in the study. Exclusion criteria were pregnant women or those menstruating during data collection. All potential respondents were screened by trained data collectors to determine the eligibility for the CKD substudy, and subsequently completed a data collection form via a face-to-face interview. The nurse in the field team was responsible for obtaining clinical measurements. Fasting blood sugar and total cholesterol were estimated by a dry method (CardioChek PA, Indianapolis, IN), which has been validated.18.Noor Ani A. Ummi Nadiah Y. Noor Azah D. et al.Sensitivity and specificity of CardioChek® PA in detecting individuals with abnormal cholesterol and glucose level.Int J Biomed. 2012; 2: 132-135Google Scholar BP was measured by Omron Japan Model HEM-907 (Tokyo, Japan), which has been validated19.Gurpreet K. Tee G.H. Karuthan C. Evaluation of the accuracy of the Omron HEM-907 blood pressure device.Med J Malaysia. 2008; 63: 239-243PubMed Google Scholar and calibrated. BP was measured with the respondent at rest in a sitting position, with the BP set and appropriate-sized cuff at chest level. Two readings were taken 15min apart, and the average measurement was used for analysis. Blood sampling for serum creatinine and early-morning urine sampling for urine ACR estimation were obtained by hemodialysis personnel from the nearest Ministry of Health hemodialysis unit. Blood and urine samples (including repeat samples) were sent by express courier to Institute for Medical Research Kuala Lumpur (central laboratory) for analysis and storage. Serum and urine creatinine and microalbumin were analyzed on Selectra XL Chemistry Analyzer (Vital Scientific, Dieren, The Netherlands) using reagents purchased from Randox Laboratories (Antrim, UK). The Jaffe method was used to measure creatinine, and has calibration traceable to an isotope dilution mass spectroscopy reference method. Interassay coefficient of variability for creatinine at 124 and 303μmol/l was 6.2% and 4.7%, respectively, and for microalbumin at 32.2 and 159mg/l was 8.0% and 3.6%, respectively. eGFR was derived for these subjects using the CKD-EPI equation5.Levey A. Stevens L. Schmid C. for the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) et al.A new equation to estimate glomerular filtration rate.Ann Intern Med. 2009; 150: 604-612Crossref PubMed Scopus (15959) Google Scholar,13.Imai E. Horio M. Watanabe T. et al.Prevalence of chronic kidney disease in the Japanese general population.Clin Exp Nephrol. 2009; 13: 621-630Crossref PubMed Scopus (382) Google Scholar as follows: 141 × min(SCr/k, 1)α × max (SCr/k, 1)–1.209 × 0.993Age × (×1.018 if female) where SCr is serum creatinine (in mg/dl), k is 0.7 for women and 0.9 for men, α is -0.329 for women and -0.411 for men, min is the minimum of SCr/k or 1, and max is the maximum of SCr/k or 1. Repeat urine measurements were obtained in respondents with albuminuria within 2 months of the initial test to estimate persistence of urine abnormalities. Patients with abnormal urine results and/or abnormal eGFR were notified to the regional nephrologist for confirmatory urine tests and further management. Microalbuminuria was defined as urine ACR 30–300mg/g. Macroalbuminuria was defined as urine ACR of ≥300mg/g.11.Wen C.P. Cheng T.Y. Tsai M.K. et al.All-cause mortality attributable to chronic kidney disease: a prospective cohort based on 462 293 adults in Taiwan.Lancet. 2008; 371: 2173-2182Abstract Full Text Full Text PDF PubMed Scopus (730) Google Scholar Persistent albuminuria was defined as two episodes of albuminuria at least 1 month apart. CKD stages 1 and 2 were defined as eGFR ≥90ml/min per 1.73m2 and 60–89ml/min per 1.73m2, respectively, with persistent urine ACR ≥30mg/g. Stages 3, 4, and 5 were defined as eGFR 30–59, 15–29, and <15ml/min per 1.73m2, respectively, regardless of kidney damage.13.Imai E. Horio M. Watanabe T. et al.Prevalence of chronic kidney disease in the Japanese general population.Clin Exp Nephrol. 2009; 13: 621-630Crossref PubMed Scopus (382) Google Scholar,16.Levey A. Coresh J. Chronic kidney disease.Lancet. 2012; 379: 165-180Abstract Full Text Full Text PDF PubMed Scopus (1293) Google Scholar Hypertension was defined as the average of two BP readings with systolic BP ≥140 and/or diastolic BP ≥90mmHg20.Chobanian A.V. Bakris G.L. Black H.R. et al.The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report.JAMA. 2003; 289: 2560-2571Crossref PubMed Scopus (16539) Google Scholar and/or self-reported hypertension previously diagnosed by medical personnel. Diabetes mellitus was defined as a fasting capillary blood glucose level ≥6.1mmol/l on CardioChek, or self-reported diabetes diagnosed by medical personnel, or random capillary blood glucose ≥11.1mmol/l. Hypercholesterolemia was defined as random or fasting blood cholesterol ≥5.2mmol/l or self-reported hypercholesterolemia diagnosed by medical personnel. Physical activity—‘inactive’ was defined as a score of <600 metabolic equivalent (MET)-min/week on IPAQ21.Institute of Public Health (IPH) 2011 National Health and Morbidity Survey 2011 (NHMS 2011), NMRR 10–757 – 6837. Volume 2: Non-communicable diseases, 188 pages, ISBN 978 967 3887 68 2.Google Scholar (International Physical Activity Questionnaire). Data analysis was done by exporting the raw data into SPSS (version 18, Chicago, IL). The data were then checked and cleaned. Distributions and categories were examined. Categories with small sample size and skewed distributions were noted. Meaningful combination of categories was done when indicated. Analysis was done according to the working definitions. Data were presented as mean and median for continuous variables and proportion for categorical variables. Prevalence estimates of all outcomes were performed. The prevalence of CKD was reported with and without correction for the persistence of albuminuria (a second sample of urine was obtained in 52.2% of initial positive readings). Factors associated with CKD were assessed using logistic regression. Unadjusted odds ratios between exposure variables and indicators of CKD were determined. A multiple logistic regression analysis was used to examine the simultaneous effects of dependent variables, that is, sex, age, ethnicity, household income group, cigarette smoking, alcohol consumption, physical inactivity, diabetes, hypertension, heart disease, family history of kidney disease, nonsteroidal anti-inflammatory drug use, and traditional medicine use with CKD status. The dependent variable is CKD status, which includes all CKD cases. A backward stepwise variable selection method was used to obtain significant variables in the model. All possible two-factor interaction terms were checked one by one together with the main effects. The statistical significance of each regression coefficient was tested using the Wald χ2 statistic. Goodness-of-fit statistics was used to assess the fit of the logistic model against the actual outcomes. Adjusted odds ratios and 95% confidence intervals were estimated. The P-value of <0.05 is considered significant. We thank the following: Director General of Health Malaysia for permission to publish this paper; Dr Zaki Morad Mohamad Zaher who initially proposed the study; Drs Liu WJ, Rosnawati Yahya, and Lim TO in helping with protocol development; Mohd Hazrin bin [email protected], Institute for Public Health for contributing Figure 1, and data collectors of NHMS 2011 for recruitment of respondents for the CKD study; Ms Nur Zati Iwani AK, Ruziana Mona WZ, and Nor Idayu R from the Institute for Medical Research for processing and analysis of all urine and blood samples; Dr Punithavathi Narayanan from Clinical Research Centre, Penang, for data handling of the blood and urine results; and Dr Tahir Aris, Head of the Institute for Public Health Malaysia, for his support. We also thank the nephrologists and paramedics of hemodialysis units (Ministry of Health) at the following hospitals for collecting urine and blood samples: Federal Territory: Kuala Lumpur Hospital, Putrajaya Hospital; Johor: Batu Pahat Hospital, Sultanah Aminah Hospital, Sultan Ismail Hospital, Kluang Hospital, Sultanah Fatimah Hospital Muar; Kedah: Baling Hospital, Sultanah Bahiyah Hospital Alor Setar, Sultan Abdul Halim Hospital Sungai Petani, Kulim Hospital; Kelantan: Raja Perempuan Zainab II Hospital Kota Bharu; Melaka: Melaka Hospital; Negeri Sembilan: Port Dickson Hospital, Jelebu Hospital; Pahang: Tengku Ampuan Afzan Hospital Kuantan, Bentong Hospital, Sultan Haji Ahmad Shah Hospital Temerloh; Penang: Penang Hospital, Bukit Mertajam Hospital, Balik Pulau Hospital; Perak: Tapah Hospital, Raja Permaisuri Bainun Hospital Ipoh, Seri Manjung Hospital, Taiping Hospital; Perlis: Sultanah Fauziah Hospital Kangar; Selangor: Selayang Hospital, Tengku Ampuan Rahimah Hospital Klang, Serdang Hospital, Kajang Hospital, Banting Hospital; Terengganu: Sultanah Nur Zahirah Hospital Kuala Terengganu, Kemaman Hospital. Last but not least, we thank all the NHMS respondents who consented to involvement in this study. The study was made possible by the National Institutes of Health, Ministry Health Malaysia, and partly supported by a research grant from the Post Graduate Renal Society Malaysia.