Leveraging Real-World Data to Conduct Noninterventional Studies in Rare Disease Populations: Lessons in Scalability and Data Access From Three Observational Studies
Author(s)
Irene Pan, MSc1, Milena Toncheva, MSc2, Miriana Wennekes, MSc2.
1United BioSource LLC (UBC), King of Prussia, PA, USA, 2United BioSource LLC (UBC), Uxbridge, United Kingdom.
1United BioSource LLC (UBC), King of Prussia, PA, USA, 2United BioSource LLC (UBC), Uxbridge, United Kingdom.
OBJECTIVES: To present the design strategies and operational challenges of non-interventional studies of ultra-rare diseases.
METHODS: We examined the design and implementation of three studies that collected real-world data (RWD) to characterise disease natural history and management. All three collected data on demographics, clinical characteristics, diagnostic history, treatment, healthcare resource utilization, and patient-reported outcomes (PRO). Recruitment strategies, methods for characterising disease progression and integrating electronic health record (EHR) data were all considered during the design phase. Cross-study challenges and solutions were compared with a major focus on scalability and data availability.
RESULTS: Key lessons common to all three studies are: (1) Well-conducted literature reviews can inform multiple aspects of the study design prior to gaining additional insights from stakeholders, (2) Stakeholder engagement including patients, clinicians and regulators is essential in shaping relevant outcomes and informing practicality of the design, (3) Early engagement of statistical experts for the selection and format of critical variables helps refine the research questions and efficiently develop the analysis plan, (4) Harmonising data standards for critical variables across sites/sources enhances data quality and interpretation of results, and (5) Flexible follow-up schedules, appropriate compensation schemes and minimal-burden PROs help reduce attrition in long-term follow-up.
CONCLUSIONS: Studies of rare disease populations typically face challenges in diagnostic heterogeneity, small sample sizes, and capturing changes in patient journey and care pathway. Methods used to address these pitfalls include innovative solutions of remote data collection and quality review; data linkages across sources via tokenisation, should be considered when applicable. The goal is to design a study that is balanced in scientific integrity and practicality. A well-designed study with a clear and streamlined data harmonisation method should be scalable both geographically and operationally to accommodate changes in real-world practice standards and evolving treatment landscapes.
METHODS: We examined the design and implementation of three studies that collected real-world data (RWD) to characterise disease natural history and management. All three collected data on demographics, clinical characteristics, diagnostic history, treatment, healthcare resource utilization, and patient-reported outcomes (PRO). Recruitment strategies, methods for characterising disease progression and integrating electronic health record (EHR) data were all considered during the design phase. Cross-study challenges and solutions were compared with a major focus on scalability and data availability.
RESULTS: Key lessons common to all three studies are: (1) Well-conducted literature reviews can inform multiple aspects of the study design prior to gaining additional insights from stakeholders, (2) Stakeholder engagement including patients, clinicians and regulators is essential in shaping relevant outcomes and informing practicality of the design, (3) Early engagement of statistical experts for the selection and format of critical variables helps refine the research questions and efficiently develop the analysis plan, (4) Harmonising data standards for critical variables across sites/sources enhances data quality and interpretation of results, and (5) Flexible follow-up schedules, appropriate compensation schemes and minimal-burden PROs help reduce attrition in long-term follow-up.
CONCLUSIONS: Studies of rare disease populations typically face challenges in diagnostic heterogeneity, small sample sizes, and capturing changes in patient journey and care pathway. Methods used to address these pitfalls include innovative solutions of remote data collection and quality review; data linkages across sources via tokenisation, should be considered when applicable. The goal is to design a study that is balanced in scientific integrity and practicality. A well-designed study with a clear and streamlined data harmonisation method should be scalable both geographically and operationally to accommodate changes in real-world practice standards and evolving treatment landscapes.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
SA62
Topic
Patient-Centered Research, Real World Data & Information Systems, Study Approaches
Topic Subcategory
Literature Review & Synthesis
Disease
Rare & Orphan Diseases