Answers are grounded in this paper and cite their sources.
Rare bone diseases (RBD) are heterogeneous conditions associated with substantial morbidity and healthcare burden. While registries enable deep phenotyping, they often lack population-level capture and insight into real-world healthcare utilization. We conducted a scoping review to evaluate how administrative claims databases have been used in RBD research and to characterize their methodological strengths and limitations. Following PRISMA-ScR guidelines, we searched PubMed/MEDLINE with complementary screening of Embase records for English-language studies using administrative or health insurance claims databases in RBD, without year restriction. The final search was completed on February 4, 2026. Data were charted using a standardized framework capturing study characteristics, database type, design, case definitions, validation procedures, and outcomes across predefined impact domains. Twenty-nine studies (2006–2026) were included, with most published since 2023 (21/29, 72%). The literature was predominantly concentrated on osteogenesis imperfecta (11/29, 38%), followed by achondroplasia and Paget’s disease (5/29 each, 17%). Most studies were conducted in the United States (17/29, 59%) and used retrospective cohort designs, frequently with matched comparator groups (16/29, 55%). Data sources included public population-based systems (15/29, 52%), private claims databases (12/29, 41%), and mixed public–private data sources (2/29, 7%). Health system utilization and costs were the most frequently examined outcomes (16/29, 55%), followed by disease burden and complications (15/29, 52%). Methodologically, most studies relied on ICD-based case ascertainment, frequently using single-code definitions (20/29, 69%), while registry linkage and external validation of case ascertainment remained uncommon (4/29, 14% each), limiting phenotype validity and comparability across studies. Claims databases provide scalable insights into healthcare utilization and disease burden in RBD but are constrained by frequent unvalidated phenotyping and limited clinical granularity. Future studies should prioritize validated and standardized case-identification algorithms, robust validation strategies, and integration with complementary real-world data sources such as registries and electronic health records.
No takes yet. Share an insight, caveat, or question.
Philippoteaux et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: