Exploring quality of life in patients with and without heart failure

• Published in: International journal of cardiology Vol. 202; pp. 676 - 684
• Main Authors: Zhang, Jufen, Hobkirk, James, Carroll, Sean, Pellicori, Pierpaolo, Clark, Andrew L, Cleland, John G.F
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ireland Ltd 01.01.2016
• Subjects: Aged; Cardiovascular; Female; Health Status; Heart Failure - psychology; Humans; Left ventricular systolic dysfunction; Male; Middle Aged; Outcome Assessment (Health Care); Patterns of quality of life; Quality of Life; Severity of Illness Index; Surveys and Questionnaires; Suspected heart failure; Patterns of quality of life; Suspected heart failure; Left ventricular systolic dysfunction
• ISSN: 0167-5273; 1874-1754
• DOI: 10.1016/j.ijcard.2015.09.076

Abstract Aims The EuroHeart Failure Survey Questionnaire (EHFSQ-1) has 39 questions on symptoms and quality of life (QoL); many items are related. We sought to identify underlying clusters amongst EHFSQ-1 questions, construct an overall “QoL score” and investigate its relationship to a single question asking patients to self-rate QoL. Methods and results Factor analysis based on the principal component technique was used to identify patterns amongst responses to QoL questions from patients referred with symptoms suggesting heart failure (HF). Of 1031 patients, median age 71 (IQR: 63–77) years, 64% were men and 626 had confirmed HF. For patients with HF, seven symptom-clusters were identified: “breathlessness”, “psychological distress”, “sleep quality”, “frailty”, “cognitive/psychomotor function”, “cough” and “chest pain”. These clusters accounted for 65% of the total variance in QoL score. Cluster pattern was similar in patients with and without HF. A summary factor score was tightly correlated with summary QoL score (correlation coefficient: r = 0.96; p < 0.0001). Both summary factors and QoL scores were highly correlated with patient self-rating of overall health (r1 = 0.61 and r2 = 0.66 respectively, p < 0.0001) or overall QoL (r1 = 0.60 and r2 = 0.66, p < 0.0001). The medians (IQR) of the summary QoL score for patients with HFrEF, HFnEF and no-HF were, respectively, 83 (60–106), 82 (59–104) and 71 (51–94). Conclusions EHFSQ-1, comprises seven symptom clusters in patients with HF. Either summary factors or QoL scores can be used as a QoL outcome measure. However, if the key question is ‘what is this patient's QoL?’ rather than the reason why it is impaired, then a single, direct question may suffice.