The Kansas City Cardiomyopathy Questionnaire (KCCQ)
The Kansas City Cardiomyopathy Questionnaire is the most sensitive, specific, and responsive health-related quality of life measure for heart failure.
The Kansas City Cardiomyopathy Questionnaire is a 23-item, self-administered instrument that quantifies physical function, symptoms (frequency, severity and recent change), social function, self-efficacy and knowledge, and quality of life.
This instrument was developed and validated by John Spertus. John is Director of Cardiovascular Education and Outcomes Research at the Mid America Heart Institute, and he is also a Professor of Medicine at the University of Missouri – Kansas City.
In the KCCQ, an overall summary score can be derived from the physical function, symptom (frequency and severity), social function and quality of life domains. For each domain, the validity, reproducibility, responsiveness and interpretability have been independently established. Scores are transformed to a range of 0-100, in which higher scores reflect better health status. For brevity, only the performance characteristics of the overall summary score are presented in this discussion.
Validity refers to the degree to which an instrument measures what it is supposed to measure. Here’s information about the validity of the KCCQ.
Since the validity of each individual domain has been independently established, all components of the summary score are considered valid representations of their intended domains. When compared with the physician-assessed NYHA, the mean KCCQ summary scores are shown in this figure:
Reliability and Responsiveness
Reliability refers to the ability of a measure to produce consistent results when the measured phenomenon is unchanged. Responsiveness refers to the ability of a measure to track accurately a phenomenon when it does change. Here’s how the KCCQ stacks up.
The test-retest reproducibility of the KCCQ was originally established in an outpatient cohort of 39 stable patients (mean age = 64 years, 69% male, mean NHYA = 2.0 ±0.59). Baseline and 3-month KCCQ overall summary scores were (66.2 vs. 64.1, mean change = -2.1, p=0.36). The intraclass correlation coefficient in a separate cohort of 320 patients from 13 centers in whom physicians re-evaluated patients 6±2 weeks apart and determined them to be stable was 0.88.
Responsiveness/Sensitivity to change
In a cohort of 39 patients (mean age =68, 62% male, mean NYHA=3.3±.46) admitted to the hospital for decompensated heart failure and re-evaluated 3 months later as outpatients (when their condition had significantly improved), the baseline and 3-month KCCQ overall summary scores were 31.8 and 56.1 (mean change = 24.3, p<0.001). In this study the responsiveness statistic (mean change/standard deviation of change in stable patients) was >2.4 times higher than the Minnesota Living with Heart Failure Questionnaire or the SF-12, suggesting greater sensitivity to clinical change than these other measures.
Several mechanisms for establishing standards for interpreting scores are available.
One is to examine the prognostic significance of KCCQ scores and the other is to benchmark score changes against clinical assessments of change. To facilitate the interpretation of cross-sectional KCCQ scores, 1,516 patients assessed 3 months after a myocardial infarction complicated by heart failure were followed for 1 year survival and heart failure hospitalization. This figure describes the Kaplan-Meier curves for this study:
To interpret changes in KCCQ scores, a cohort of 460 patients from 13 centers were followed for 6±2 weeks at which point physicians assessed their clinical change (blinded to KCCQ scores). Among these patients, the magnitude and direction of change was as follows: large deterioration, n=5 (1%); moderate deterioration, n=13 (3%); small deterioration, n=35 (7%); no change, n=320 (67%); small improvement, n=65 (14%); moderate improvement, n=34 (7%); and large improvement, n=4 (1%). The KCCQ change scores were exquisitely reflective of clinical changes in heart failure both in terms of its directionality (improvement versus deterioration) and proportion-al-ity of change (magnitude) — as revealed in this figure:
For patients experiencing large, moderate and small deteriorations in their condition, KCCQ Overall Summary scores decreased by -24.9±15.8, -16.8±13.8 and -5.4±10.8 points. For those with no, small, moderate and large improvements in their heart failure, the KCCQ scores improved by 1.3±11.8, 5.7±16.1, 10.7±16.2 and 22.3±15.0 points. The mean change in KCCQ scores was significantly different for all categories of change compared to stable patients. Even those with small clinical deteriorations or improvements (-5.3±11 for small deterioration vs. 1.3±12 for no change, p=0.002 and +5.7±16 for small improvement, p=0.01). This suggests that a mean difference over time of 5 points on the KCCQ Overall Summary Scale reflects a clinically significant change in heart failure status.
An alternative approach to interpreting clinical changes is to appreciate the prognostic significance of changes in scores. In a cohort 659 subjects assessed 3 and 6 months after an MI complicated by CHF, those patients whose KCCQ overall summary scores declined by ≥ 10 points had a 107% increased risk of dying or being hospitalized over the next 3 months (event rate = 11.4% vs. 5.5%,p<0.001). By one year after the last KCCQ assessment, a 94% increased relative risk of cardiovascular death/hospitalization was observed (27.7% vs. 14.3%, p<0.001).
These data suggest that a 10 point decline in KCCQ scores has important prognostic significance.
Created: April 29, 2004 03:34
Last updated: December 16, 2015 23:55