Monitoring diabetes is critical for our understanding of the etiology and natural history of disease and for public health actions. However, traditional methods for monitoring are either too expensive (e.g., IDDM registries, NIDDM-OGTT prevalence surveys) or too inaccurate (routinely collected data or passive surveillance) for broad accurate, national programs for monitoring the incidence and prevalence of disease. We suggest that one technology called capture-recapture would considerably increase our ability to “count” diabetes, both nationally and globally. Implementation of this approach could lead to accurate inter- and intracountry data on rates of disease. Moreover, such tracking of diabetes could serve as the model for the monitoring of all disease in the 21st century and beyond.