Administrative insurance claims are the backbone of real-world evidence (RWE). However, for those of us working in the rare and ultra-rare disease space, traditional "closed" claims often feel like trying to view a landscape through a keyhole.
The bottleneck is the requirement for continuous enrollment. In most cases, losing a few patients due to a change in insurance is a rounding error. In rare diseases, it’s a data catastrophe. When we restrict a study only to patients who stay with a single payer, we aren't just losing numbers, we’re losing the full story.
Hybrid claims data, which integrates both open and closed sources, offers a solution. By looking beyond the rigid boundaries of a single insurance plan, we can significantly boost cohort sizes and, more importantly, create evidence that actually reflects the real world.
This blog is part of a series on hybrid claims data and real-world evidence. Read the first post here.
In the ultra-rare world, every patient is a data point we can't afford to lose. Yet, traditional methodology often acts as a sieve.
Take a recent use case involving a pediatric ultra-rare treatment. We started with 451 patients who had at least three paid claims for the therapy. But once we applied the standard "closed claims" filter, the group withered:
For an epidemiologist, this is where a study hits a wall. When your sample size is this small, statistical power evaporates. You can’t do meaningful subgroup analyses, and your confidence intervals become so wide they’re practically unusable for HEOR.
Now, consider the shift when we introduce open claims. These follow the patient regardless of their payer, capturing the "breadcrumbs" of their journey across the healthcare system. In that same pediatric study, the numbers changed overnight:
By combining these into a hybrid cohort, the number of viable patients for a full longitudinal study jumped from 109 to 220. We didn’t just "add more rows" to a spreadsheet; we doubled the statistical power of the study.
This isn't just a numbers game. It’s about generalizability. Closed claims are inherently biased toward "stable" patients. They change jobs, switch from commercial insurance to Medicaid, or age into different coverage tiers.
When we only study the patients who stay put, we ignore the high-need populations who experience coverage gaps or provider transitions. Hybrid data reduces this selection bias, offering:
1. Broader Geography: Capturing patients in regions where a single payer might not be dominant.Increasing sample size does three critical things:
In rare disease HEOR studies, longitudinal depth is often as important as cohort size.
In the ultra-rare example, hybrid integration enabled more patients to meet longitudinal criteria, strengthening outcome measurement credibility and the overall defensibility of the evidence. This creates insights that are more useful for long-term decision making.
The ability to say, “This analysis reflects a broader, more representative treated population” is powerful.
Moving from Incremental to Foundational
For biopharma teams, the implications are practical. A study that looks "underpowered" on paper using traditional data might actually be viable in a hybrid environment. More importantly, the evidence becomes more defensible. When you present a value dossier to a regulator or a payer, being able to say your data represents the actual treated population, not just a sliver of one insurer's database, is a massive advantage.
The industry is moving past the era where closed claims alone are enough. In rare disease research, where the patient count is the ultimate constraint, hybrid integration isn't just a luxury. It’s the new standard for building evidence that sticks.