Computer Adaptive Test for Patients with Heart Failure

DESCRIPTION (provided by applicant): Our objective is to achieve a major advance in the technology used to assess patient reported outcomes in heart failure (HF). Numerous questionnaires currently are being used to measure outcomes of HF patients, but scores from these instruments can not be directly compared. In addition, no single fixed-length form can achieve the breath and precision of measurement needed to monitor individual patients, and yet still be practical for clinical use. To address these deficiencies, the Aims of Phase I are: (1) to develop a prototype Computer Adaptive Test (HF-CAT) to measure the cardinal subjective symptoms of heart failure (dyspnea, fatigue and physical function); (2) to develop a cross-calibration-approach to explore the possibility of comparing disease-specific static questionnaires (Chronic Heart Failure, Minnesota Living with Heart Failure, and Kansas City Cardiomyopathy Questionnaire); and (3) to evaluate the feasibility of a HF-CAT. We will collect data via the internet from 1,500 HF patients, who will answer the instruments mentioned above, two dyspnea questionnaires, and the vitality and physical functioning scale from the SF- 36. Item Response Theory methods will be applied to estimate item parameters to extend existing item banks for fatigue and physical functioning and to build a new item bank for dyspnea. These item banks will serve as a basis for the HF-CAT and will enable us to explore the possibility of comparing results between questionnaires. We will evaluate measurement precision and respondent burden in two simulation studies, and evaluate the feasibility of the HF-CAT in a study with 100 HF patients at the Montefiore Medical Center. Phase I will produce a prototype version of the HF-CAT with preliminary evidence of its acceptance by patients and a better understanding of the shared concepts of widely-used HF questionnaires. We expect the HF-CAT will provide enhanced measurement precision over a wide scale range without added respondent burden. Enhanced measurement precision is a desired feature for psychometric instruments, to improve response detection and risk prediction. In Phase II, we will include additional items, examine the responsiveness of the HF-CAT, and conduct clinical field studies to evaluate the tool's ability to predict deterioration of HF and health care utilization. The final goal will be to have a comprehensive practical tool available to improve patient centered care and life expectancy of patients with heart failure. Public Heath Relevance: Within this project we will develop a computer adaptive test which will measure the subjective symptoms of heart failure more precisely and with less respondent burden. This may help high risk heart failure patients get appropriate treatment earlier to prevent deterioration and hospitalization.