Comparative reproducibility and validity of systems for assessing cardiovascular functional class: advantages of a new specific activity scale.

Reproducibility and validity are prerequisites for a useful clinical scale. We therefore prospectively tested the reproducibility and validity of the New York Heart Association criteria and the Canadian Cardiovascular Society criteria for the assessment of cardiac functional class and compared these criteria with a new Specific Activity Scale based on the metabolic costs of specific activities. The New York Heart Association estimates made by two physicians had a reproducibility of only 56%, and only 51% of the estimates agreed with treadmill exercise performance. Functional estimates based on the Canadian Cardiovascular Society criteria were significantly more reproducible (73%), but not significantly more valid. The Specific Activity Scale was as reproducible as the Canadian Cardiovascular Society criteria, and its 68% validity was significantly higher than the validities of the other systems. The easily administered Specific Activity Scale was equally reproducible and valid when used by a nonphysician. It was especially better than the other systems for the evaluation of true class II patients and was significantly less likely to underestimate treadmill performance. Although no set of questions can perfectly predict exercise tolerance, the Specific Activity Scale deserves wider prospective testing.