Selection of bolt-ons after factor analysis identification: are linear regression models a useful technique?
posterposted on 25.02.2020 by Aureliano Paolo Finch, John Brazier, Clara Mukuria
Poster sessions are particularly prominent at academic conferences. Posters are usually one frame of a powerpoint (or similar) presentation and are represented at full resolution to make them zoomable.
It is now recognized that the EQ-5D may miss dimensions important for some conditions. When this happens, a possible solution is adding bolt-ons to expand its descriptive system. Previous bolt-on studies have identified potential candidates using information on validity in specific areas such as vision (1). Although this is a useful approach for identifying individual bolt-ons, it does not help in identifying what other dimensions may be missing from the EQ-5D. Factor analysis has been seen to be a potential approach for bolt-on identification. This techniques pinpoints to a list of factors, and items loading on them, that are not related to the EQ-5D latent constructs(2). These can be adapted / developed into bolt-ons. However, not all bolt-ons can be added to the EQ-5D simultaneously, as this would affect the measure’s acceptability and feasibility. Hence, methods to select bolt-ons from the identified list are needed. This study investigates the possibility of using linear regression models for the selection of bolt-ons after factor analytic identification
EthicsThere is no personal data or any that requires ethical approval
PolicyThe data complies with the institution and funders' policies on access and sharing
Sharing and access restrictionsThe data can be shared openly
- The file formats are open or commonly used
Methodology, headings and units
- Headings and units are explained in the files