There are challenges associated with generating health-state utility values for rare diseases, leading to a potential lack of standardization in the methods used. This systematic literature review characterized the approaches used to generate utility data in rare diseases.
Searches of MEDLINE/Embase, health technology assessment and cost-effectiveness databases were conducted, supplemented by grey literature searches. Due to the large volume of evidence identified, articles were prioritized for full-text review by applying a 2020 date limit.
Ninety-seven articles (assessing 56 rare conditions) were included. Nineteen unique health-related quality-of-life tools were identified, 14 of which were generic. Indirect valuation methods were more common than direct (80 vs 43 instances). Among the indirect methods, the preference-based EQ-5D questionnaire was most reported (55 instances), followed by the non-preference-based short-form questionnaires (8 instances). Five disease-specific, non-preference-based questionnaires were reported. Mapping algorithms were used for preference-based and non-preference-based measures, typically mapped to EQ-5D, although challenges with mapping disease-specific tools to preference-based measures were noted. Vignettes were used in 29 articles; however, incomplete reporting on the development process limited the quality assessment.
Generic, preference-based measures were commonly used to generate utility data in rare diseases among the identified studies, facilitating comparison but potentially limiting sensitivity of results. Development of appropriate and valid disease-specific measures and more transparent/consistent reporting of vignette development, would help ensure that all aspects of health-related quality-of-life impacted by rare diseases are suitably captured, to allow reliable demonstration of the value of treatments to support future reimbursement.