Little effort has been directed toward studying the relationship between morbidity and mortality at exceptional ages, perhaps for no better reason than it has been difficult to do given available data resources. Two study innovations/adjustments are required to adequately represent count data with reports of health from aged sample persons. These design features are oversampling of the exceptional group and linkage to detailed administrative reports. The National Long-Term Care Survey (NLTCS) has made it possible to study health and functioning in the context of exceptional longevity.