Diabetic nephropathy was first described in patients with non-insulin-dependent diabetes mellitus (NIDDM, type II diabetes) by Kimmelstiel and Wilson in 1936. It is a classical, late diabetic complication, diagnosed by measurement of the urinary albumin or total protein excretion, and typically develops after more than 15 years of diabetes. As most studies of patients with type II diabetes have been performed in White, European or North American populations in which the highest incidence of this disease is recorded in individuals aged over 70 years, a low prevalence has generally been found in these patients. Nephropathy has been considered a rare complication in type II diabetes patients. Other ethnic groups such as Pima Indians in the USA or Pacific Islanders have totally different incidence patterns of type II diabetes, with a high incidence in the 20- to 50-year age group. These patients live long enough to develop nephropathy, and they do so at the same rate as insulin-dependent diabetes mellitus (IDDM, type I diabetes) patients. Since the prevalence of type II diabetes is increasing worldwide, particularly in the developing world, diabetic nephropathy will be a growing problem in patients with this disease. From our experience in the treatment of type I diabetes patients, we know that prevention of end-stage renal failure is possible in most patients, but that treatment of end-stage renal disease is very expensive. In this paper, some of the major risk factors for the development of nephropathy are discussed together with the potential for treatment. It is shown that, in type I diabetes patients, early detection by screening for microalbuminuria and immediate recourse to antihypertensive treatment are likely to increase life-expectancy significantly while at the same time reducing the total costs to healthcare. Whether this is also the case in patients with type II diabetes is less clear, as most of the controlled clinical trials of the effect of strict metabolic control or antihypertensive treatment have been performed in patients with type I diabetes. Thus, clinical trials in patients with type II diabetes should be performed, and further epidemiological data relating to these patients are needed.