Capturing individual experiences: advancing core outcome sets in pain. Comment on Taylor et al. (2024)

Résumé fourni par la source

Core outcome sets1 are increasingly developed to assess what is relevant to researchers, clinicians, and patients in intervention studies. A challenge remains in measuring and analysing these outcome domains with specific instruments and in using these assessments to guide personalized treatment. Recognizing that everyone experiences pain differently, Taylor et al.4 propose a holistic minimal clinically important difference (MCID) across 5 core domains: pain intensity, physical function, health-related quality of life, emotional function, and sleep quality. Interestingly, the MCID is personalised by averaging across the domains where individual's scores were considered impaired based upon clinical cut-offs. Taylor et al. demonstrate the internal and construct validity of the MCID to inform its use in treatment trials. While this personalisation is a step forward, some concerns remain. Taylor et al. calculated the MCID for baseline domains based on clinical cut-offs without directly asking patients about the domains that are relevant in their life. The meaning and value individuals attribute to core outcomes may be highly individual.2 A recent focus group with people with chronic pain suggested that QoL may mean something different for every individual, and that the meaning they attribute to it may even fluctuate over time (Pogatzki-Zahn, Esther M., De Paepe, Annick L. & Crombez, Geert, unpublished study (focus group conducted in May 2024). The best approach is then to ask participants about their most important domains and assess core outcomes based on these. In addition, further steps are needed to truly personalize. Items from most measures are selected using item validation procedures and statistics that optimize relations at a group level. Such “nomothetic” measures do not guarantee that each item is relevant for each patient. So, asking only which domains are of value in the life of patients ignores the fact that some items to assess the valued domain may be irrelevant for an individual patient. The EQ-5D, used by Taylor, clearly illustrates this point. The EQ-5D includes items on mobility, self-care, usual activities, pain/discomfort, and depression/anxiety. An individual may score low on one critical item, affecting their QoL, but high on others. A mean score may not accurately reflect their QoL. A more precise and sensitive-to-change measure could be obtained from an individualized, or “idiographic,” questionnaire omitting items that are irrelevant for a particular person.3 In creating these measures, it is crucial to start by assessing content validity at the individual level (ie, what the items measure for a particular person). In fact, content validity is the most important measurement property for a patient-reported outcome.5 In conclusion, while Taylor et al.'s approach to personalizing the MCID across core domains is a significant advancement, it is essential to further refine this method to truly capture individual patient experiences. By directly involving patients in identifying the domains most relevant to their lives and ensuring that assessment tools are tailored to these individual priorities by assessing their content validity, we can enhance the accuracy and relevance of outcome measures. Disclosures The authors have no conflict of interest to declare.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.