fact, they estimated that 5.9 million have stage 1 CKD with normal renal function and GFR >90 mL/min, whereas another 5.3 million have stage 2 (GFR 89–60 mL/min) [9]. Studies in other Western countries show that around 6% to 7% of the population has albuminuria, with around 0.6% to 0.7% overt proteinuria [10]. In Australia, the AusDiab study showed that 10% of those screened had impaired renal function; 16% of those had some degree of kidney involvement, impaired renal function, hematuria, and/or proteinuria [11]. Studies in developing countries such as Singapore showed a similar prevalence of CKD [12]. Before we become alarmed by these findings and assume that millions of individuals worldwide are heading toward RRT, we need to be somewhat critical of these analyses and raise some issues. The NHANES III data were cross-sectional in nature, with individuals tested once. The variability of measurements of albuminuria/proteinuria, as well as those based on serum creatinine estimation, is quite high, and rendering definitive conclusions based on such cross-sectional analysis somewhat doubtful. Albuminuria/proteinuria, as we all know, can be affected by a range of factors, including posture, timing of the sample, exercise, infection, and pyrexia. Also, studies have shown that with repeat testing of a random population, only 60% remain positive [9, 13, 14]. Similarly, measurements of serum creatinine require standardized calibration across laboratories, as variation in a given sample estimate can reach 25% [14]. The calculation of glomerular filtration rate (GFR) based on the MDRD formula can add additional bias to the estimation of the number of patients with different stages of CKD, as its accuracy in the estimation of GFR has not been validated for general screening, for testing individuals with normal renal function, or in ethnic minorities [15]. In fact, recent data suggest that Hispanic Americans, particularly those of Cuban descent, have higher serum creatinine levels when compared to whites [16]. Whether this discrepancy reflects a higher prevalence of CKD in this ethnic minority or merely differences in body mass and creatinine handling is unknown, but the finding highlights difficulties with estimation of renal function based on serum creatinine levels. Also, analysis of individuals with reduced GFR does not always take into account age-related changes; it is unlikely that elderly individuals with a mildly reduced GFR would have the same prognosis as patients with chronic glomerulonephritis with hypertension, proteinuria, and declining renal function. A cross-sectional estimation of renal function does not provide insights into the progressive nature of CKD. Debates should be informed by more research and epidemiologic data, including longitudinal studies. This is particularly relevant to developing countries, where resources are scarce and priorities in health care must be carefully justified. Care for patients reaching ESRD and requiring RRT is costly. In the US, the current annual cost of $17 billion is expected to climb to $29 billion by 2010 [2, 17]. In Europe, dialysis alone consumes about 2% of the health care budget, with only a small fraction (2 units/day) has been linked in one survey to the development of CKD. The role of factors such as chronic consumption of analgesics and nonsteroidal antiinflammatory drugs remains debatable. Also, herbal remedies have been implicated in the development of chronic interstitial renal disease. Environmental pollution and exposure to heavy metals, including lead, have been associated with a higher prevalence of CKD [7]. Finally, there is little doubt that many of the communicable, infectious diseases affecting the developing world such as malaria, schistosomiasis, hepatitis C, and HIV increase the risk of developing CKD. In established CKD, a faster rate of progression has been associated with systemic hypertension, proteinuria, hyperlipidemia, smoking, and obesity (reviewed in [22]). Recently, K/DOQI put forward a new classification of CKD, with 5 stages based on the level of renal functional impairment [23]. This classification has been enthusiastically adopted by most. While useful in simplifying the categorization of CKD, this classification has its limitations. Stage 1 CKD, for instance, allows for the inclusion of individuals with minimal renal functional and/or radiologic abnormalities that might prove of little long-term clinical relevance. To categorize those individuals as suffering from CKD might be misguided in more than one respect, not the least psychologically [24]. It would be more appropriate to divide patients into those with some defined abnormalities, such as isolated hematuria or microalbuminuria, and those with impaired renal function. The latter might warrant sub-classifications based on the presence or absence of progression and the associated risk factors such as hypertension and proteinuria [24]. The K/DOQI classification has been of good administrative and educational value in simplifying CKD and generating a standardized global definition. It will undoubtedly undergo revisions with time that will render it more and more accurate and clinically relevant. In fact, it already has been the subject of a recent position statement from Kidney Disease: Improving Global Outcomes (KDIGO) [25]. Having defined the stages of CKD, it is imperative that we appreciate