Just a moment, the page is loading...

Statistical Methods for Analyzing KCCQ in Heart Failure Trials

Statistical Methods for Analyzing KCCQ in Heart Failure Trials

DoHwan Park, Ph.D.

Department of Mathematics & Statistics University of Maryland Baltimore County Baltimore, MD

24 Oct 2019

Chronic heart failure (CHF), either with reduced ejection fraction or preserved ejection fraction, is associated with symptoms that have an enormous impact on patients' daily lives. Improving CHF patients' quality of life is an important treatment goal, in addition to preventing disease progression with respect to mortality and hospitalization. Kansas City Cardiomyopathy Questionnaire (KCCQ) is a HF-specific 23-item self-administered questionnaire developed to evaluate HF patients' health-related quality of life. Specifically, KCCQ domains include physical limitation, symptoms (frequency, severity, and change over time), quality of life, social limitations, and self-efficacy. Some recent clinical studies have adopted KCCQ to measure effect of drug or non-drug treatment on quality of life. Given the relatively high mortality rate in the heart failure population, a major challenge in the statistical analysis of KCCQ concerns missing data due to death. Conventional statistical analysis of KCCQ such as linear mixed models for repeated measures may not reflect the underlying treatment effect, especially when the treatment has an impact on mortality or other censoring events. There is a need for novel statistical methods to better estimate and characterize the treatment effect on KCCQ while accounting for death or other potentially informative censoring events. The methodology development is relevant for the quality of life instruments in the heart failure field as well as other fields with similar challenges such as clinical research in cancer. The objectives of this research include evaluation of the existing statistical methods and development of new statistical methods as applied to KCCQ, with a particular focus on joint modelling of longitudinal and time-to-event data. In order to evaluate, develop and test statistical methods, it is important to have access to patient level data sets from contemporary, sufficiently large, and well conducted randomized clinical trials such as PARADIGM-HF.

[{ "PostingID": 4062, "Title": "NOVARTIS-CLCZ696B2314", "Description": "A multicenter, randomized, double-blind, parallel group, active-controlled study to evaluate the efficacy and safety of LCZ696 compared to enalapril on morbidity and mortality in patients with chronic heart failure and reduced ejection fraction" }]

Statistical Analysis Plan