Plain Language Summary

What is it about?

Cognition is the most commonly used additional dimension (bolt-on) for the EQ-5D health questionnaire, with many different versions varying by descriptors and response levels. This topic is important because the EQ-5D is widely used to measure health-related quality of life but may not fully capture cognitive aspects that are relevant for many patient groups. The researchers specifically aimed to determine which cognition bolt-ons demonstrate the strongest psychometric properties when added to the EQ-5D-3L (3-level) and EQ-5D-5L (5-level) versions. This study fills a knowledge gap by systematically evaluating the measurement properties of various cognition bolt-ons across different populations and contexts. The authors offer a comprehensive assessment of 28 different cognition bolt-ons to help identify which versions might be most appropriate for different applications. This systematic review makes a central contribution by synthesizing evidence from over 100 publications to guide future development and selection of cognition bolt-ons.

How was the research conducted?

The researchers used a systematic review methodology to identify, evaluate, and synthesize all available evidence on the psychometric properties of cognition bolt-ons. This approach was applied by searching three major databases (PubMed, Web of Science, and Google Scholar) following PRISMA 2020 guidelines to identify relevant studies examining cognition bolt-ons for the EQ-5D. The researchers extracted data from 101 publications representing 72 unique studies that evaluated at least 1 psychometric property of cognition bolt-ons in populations aged 16 years or older. They assessed multiple psychometric properties including ceiling effects, floor effects, informativity, convergent validity, divergent validity, known-groups validity, structural validity, responsiveness, test-retest reliability, and patient-proxy agreement. The researchers developed a specific checklist to evaluate each bolt-on version, rating psychometric properties as either positive or nonpositive, allowing comparison across different bolt-on versions and populations.

What were the results?

The main finding was that 5-level cognition bolt-ons generally outperformed 3-level versions based on the positive-to-nonpositive ratio of psychometric assessments (5L: 45/28; 3L: 55/57). The most frequently reported psychometric properties were item-level ceiling effects (75 publications) and known-groups validity (54 publications), while fewer studies examined convergent validity, divergent validity, responsiveness, patient-proxy agreement, and test-retest reliability. Notably, none of the included studies assessed content validity, which is a critical gap in the evidence base. Several bolt-ons demonstrated good divergent validity from the core EQ-5D dimensions and adequate known-groups validity for relevant clinical populations such as those with head/brain injuries or dementia. The most extensively tested bolt-ons were the Janssen 2013 version (5L, cognition: 18/15) and the Haagsma 2005 version (3L, thinking ability: 8/12), though the evidence remains insufficient to identify a clearly superior cognition descriptor.

Why are the results important?

These results have specific significance for improving health-related quality of life measurement in populations where cognitive functioning is an important aspect of health. The findings might contribute to changing clinical practice by informing the selection of appropriate cognition bolt-ons for specific patient populations, particularly those with neurological conditions, dementia, or traumatic brain injuries. Researchers, healthcare decision makers, and patients will benefit from these findings through more a comprehensive assessment of health status that includes cognitive functioning, potentially leading to better healthcare resource allocation and treatment decisions. In the long term, these results could lead to the development of standardized cognition bolt-ons that are validated across multiple languages and contexts, enabling more accurate health technology assessments and cross-national comparisons of health interventions affecting cognitive function.

What are the strengths and weaknesses of this study?

The main strength of this study is its comprehensive scope, synthesizing evidence from more than 100 publications across 18 countries to provide a thorough evaluation of cognition bolt-ons. The primary weakness is the lack of content validity evidence for any of the bolt-on items, making it difficult to determine whether they adequately capture the concept of cognition from patients' perspectives. Future research should prioritize qualitative testing of promising candidate wordings to establish content validity before conducting quantitative psychometric testing, ideally including translatability assessments to ensure cross-cultural equivalence of cognition bolt-ons across different languages and contexts.

Note: This content was created with assistance from artificial intelligence (AI) and has been reviewed and edited by ISPOR staff. For more information or for inquiries on ISPOR’s AI policy, click here or contact us at info@ispor.org.