Relationship between generic instruments used to measure healthrelated quality of life same or different

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): The Swedish research Council for Medicine and health - VR Swedish Research Council for health, working life and welfare - FORTE Background Health-related quality of life (HRQoL) is one and possibly the most important follow-up measure that is used to assess outcomes after long term and critical illness, such as heart failure. There is today an abundance of available generic measures relevant for all types of diseases and health states, and therefore suitable in heterogenous patient cohorts. Globally, two of the most common generic measures of HRQoL in use today are the EuroQol-5D (EQ-5D) and the RAND-36/36-item short-form SF-36. Another commonly used instrument is Cantril’s ladder of life. With the purpose to create a valid but not burdensome survey before and after an intervention in a cohort of former critically ill patients after a period of cardiac care, the question was raised what instrument to choose, and if the instruments correlates. If so, not all items have to be used if they measure the same outcome. Purpose To examine the relationship between different generic instruments for HRQoL in patients with heart failure. Methods For EQ-5D the separate five separate dimensions, the index from the combined dimensions, and a visual analog scale (VAS) can be used. For RAND-36 the eight dimensions created from 35 of the 36 questions, and the extern item as a single item called "Health Transition" can be used. Finally, for the Ladder of life the three questions can be used as separate items. The items chosen and measured in a cohort of former critically ill cardiac patients (n=246) were EQ-5D VAS, EQ-index, RAND-36 health transition question, and ladder of life question 1 and 3. Higher scores indicates higher perceived HRQoL on the scales except for RAND-36. The patients had a mean age of 66 years (SD 11.8), 67.3% (n=165) were men. The most common diagnoses in the sample were heart failure 68% (n=148), diabetes 22% (n=31), cancer (17% (n=36), and pulmonary disease 14% (n=31). The Spearman´s rank correlation was used to analyze the strength of association in the nonparametric data. Results Up to moderately strong correlation was seen between the items (see table). The strongest correlation was found between the EQ-5D VAS and RAND-36 Health Transition item (-0,629 (p=0.01)), followed by both RAND-36 and EQ-5D VAS and the ladder of life question "Where on the ladder do you feel you personally stand right now" (0,615 (p=0.01)) and (0.606 (p=0.01)) respectively. The two ladder of life questions correlated moderately (0.689, (p=0.01)). Conclusions Albeit the same focus area, the chosen general instruments, and items for measuring HRQoL was not strongly correlated indicated that they do not measure the exact same concepts and dimension. This is of importance when chosen instruments for HRQoL follow-up. There might be a need to combine more than one generic instrument considering the balance between the burden for the patient and the purpose and depth of the investigation. Further conceptual consideration regarding to these different generic instruments are warranted.