Back

Protocol for Generation of a Patient-Reported Outcome Measure of Quality of Life in Heart Valve Disease: The VALVQ

Pons, A.; Whalley, G.; Jenkinson, C.; Morley, D.; Coffey, S.

2023-05-21 cardiovascular medicine
10.1101/2023.05.20.23290285 medRxiv
Show abstract

BackgroundThere is an increasing prevalence of people worldwide with heart valve diseases (HVD), especially rheumatic heart disease, aortic stenosis, and mitral regurgitation, as well as people with a previous valve repair or replacement. Treatment decisions for HVD can be complex, making quality of life an important factor, but no questionnaire to measure quality of life across the lifespan of HVD exists. In this article, we describe the protocol for the development of such a questionnaire. Methods and ResultsThe project will occur over four phases. First, people with HVD, family members and clinical experts will be interviewed to generate a list of questions ( items) that comprehensively describe participants quality of life. In the second phase, this will be formatted into a questionnaire that is pilot tested for functionality. In the third phase, items will be selected according to item distributions, factor analysis and rotation, and item response theory using the Graded Response Model to generate a final questionnaire containing only the best-performing items, which will then be tested for validity. Validity assessments will be repeated after final questionnaire administration in a new sample in the fourth phase. ConclusionThe article gives a template for development of a patient report outcome measure (PROM) in the health sciences. It is expected that the final questionnaire, called the VALVQ, will allow clinical trials to more sensitively assess quality of life changes across the spectrum and lifespan in HVD.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.