J Cox1, C D Naylor. 1. Sunnybrook Health Science Centre, Toronto, Ontario, Canada.
Abstract
OBJECTIVE: To appraise the measurement properties of the Canadian Cardiovascular Society (CCS) classification of stable angina pectoris. DATA SOURCES: Relevant articles were identified through a MEDLINE search (1976 to November 1991). Bibliographies of retrieved articles were also reviewed. STUDY SELECTION: Studies chosen directly addressed the validity and reliability of the CCS scale. Recent studies and reviews of related topics (for example, silent ischemia) are selectively cited. DATA SYNTHESIS: No data address the scale's applicability, that is, how clinicians typically assign angina grades in practice. Comprehensiveness would be improved by coverage of the patient's perceptions of symptom burden; mixed exertional and rest symptoms; episodic or changing symptoms; and modifying factors. Reliability was assessed in one study with two clinicians; the interobserver, chance-corrected agreement on patient grading was 60%. Content validity (the ability of the scale to measure what it claims) is threatened by the unproven assumption of symptomatic or physiologic equivalence among diverse levels of different activities within any given grade of angina. Construct validity is uncertain, given weak relations between angina grade and noninvasive markers of ischemia, anatomical disease, or prognosis. The scale's responsiveness (the ability to detect the smallest clinically important changes) is limited by the reliance on four coarse gradations based on only ambulation or stair-climbing. CONCLUSIONS: The CCS scale for stable angina might be made more useful by developing measurements for patients' self-rated symptom burden and the changes they deem important; by adding items on clinical instability (that is, progressive symptoms or pain at rest); and by empirically testing the current scale to eliminate redundant or inconsistent elements.
OBJECTIVE: To appraise the measurement properties of the Canadian Cardiovascular Society (CCS) classification of stable angina pectoris. DATA SOURCES: Relevant articles were identified through a MEDLINE search (1976 to November 1991). Bibliographies of retrieved articles were also reviewed. STUDY SELECTION: Studies chosen directly addressed the validity and reliability of the CCS scale. Recent studies and reviews of related topics (for example, silent ischemia) are selectively cited. DATA SYNTHESIS: No data address the scale's applicability, that is, how clinicians typically assign angina grades in practice. Comprehensiveness would be improved by coverage of the patient's perceptions of symptom burden; mixed exertional and rest symptoms; episodic or changing symptoms; and modifying factors. Reliability was assessed in one study with two clinicians; the interobserver, chance-corrected agreement on patient grading was 60%. Content validity (the ability of the scale to measure what it claims) is threatened by the unproven assumption of symptomatic or physiologic equivalence among diverse levels of different activities within any given grade of angina. Construct validity is uncertain, given weak relations between angina grade and noninvasive markers of ischemia, anatomical disease, or prognosis. The scale's responsiveness (the ability to detect the smallest clinically important changes) is limited by the reliance on four coarse gradations based on only ambulation or stair-climbing. CONCLUSIONS: The CCS scale for stable angina might be made more useful by developing measurements for patients' self-rated symptom burden and the changes they deem important; by adding items on clinical instability (that is, progressive symptoms or pain at rest); and by empirically testing the current scale to eliminate redundant or inconsistent elements.