In contrast to the associations of high body mass index (BMI) with increased mortality in the general population, high BMI is associated with better survival in dialysis patients. Nonetheless, high BMI/adiposity in chronic kidney disease (CKD)/dialysis patients is associated with insulin resistance, inflammation, dyslipidemia, atherosclerosis and coronary calcification as described in the general population. These apparently perplexing associations might be explained if (1) adiposity has dual competing effects on survival; a protective nutritional effect and a deleterious metabolic effect resulting in insulin resistance, dyslipidemia, hypertension and inflammation and (2) the level of kidney function modifies the relative importance of these effects. In this paradigm, the deleterious metabolic effects of obesity outweigh its protective nutritional effects in the non-CKD population, the deleterious metabolic effects of obesity are neutralized by its protective nutritional effects in the moderate CKD population and the deleterious metabolic effects of obesity are outweighed by its protective nutritional effects in stage V CKD on dialysis. In other words, the over-all effects of obesity on survival vary according to the level of kidney function and there is an interaction of body size and presence or absence of CKD on survival even though the metabolic effects of adiposity are not modified by the level of kidney function. Therefore, we propose that despite an association of adiposity with better survival, there is no reverse epidemiology of the associations traditional and nontraditional cardiovascular risk factors and disease with adiposity in dialysis patients.