As real‑world data (RWD) becomes a central pillar of evidence generation, the ability to integrate high‑quality datasets has never been more important. Healthcare organizations increasingly rely on linked RWD to understand patient trajectories, treatment response, and clinical and economic outcomes in ways that single data sources may not support.
A good example of a condition where integrating data from disparate data sources has tremendous value is lung cancer. Administrative coding for lung cancer does not provide details about subtype, stage, and other prognostic factors. Being able to integrate additional clinical features into a RWD analysis of lung cancer is key as lung cancer continues to be the leading cause of cancer death in the U.S., and more than 50% of patients are not diagnosed until their disease has metastasized. MarketScan and Biodesix worked together to create a custom data linking project to conduct research to link MarketScan administrative claims data and Biodesix’s data with results from their blood-based immune profiling diagnostic test.
The results were discussed in a recent webinar with MarketScan and Biodesix experts, where they demonstrated how rigorously linked data can illuminate insights that would otherwise remain inaccessible.
The VeriStrat® test: Why clinical diagnostic data must be linked
VeriStrat®, one of the first proteomic tests used in lung cancer, categorizes patients into “Good” or “Poor,” providing oncologists with actionable guidance on who is less likely to respond to standard therapies. While clinical trials validate its performance, they cannot fully capture how test classifications influence real‑world treatment pathways, outcomes, or utilization patterns. Linked RWD bridges this gap by showing whether the prognostic information observed in controlled settings holds true across diverse clinical environments.
Why MarketScan + Biodesix needed to link their data
No single dataset tells the whole patient story. Biodesix’s lab data includes rich diagnostic information but lacks longitudinal insights into treatment sequencing, disease progression, or healthcare utilization. Conversely, MarketScan’s closed claims data offers deep visibility into real‑world costs, service patterns, and outcomes, but does not capture diagnostic classifications. When kept separate, each tells only part of the story. Linking them allows researchers to evaluate how proteomic classifications correlate with real‑world trajectories—precisely the type of evidence stakeholders increasingly expect.
Key research questions enabled by data linking
By integrating diagnostic results with longitudinal claims, researchers were able to investigate questions that cannot be answered using either dataset in isolation. For example:
-
Do patients classified as VeriStrat “Poor” follow different treatment pathways than those classified as “Good”?
-
Are there measurable differences in healthcare utilization across result categories?
-
How well do real‑world treatment decisions align with the risk stratification suggested by the test?
These questions reflect a broader trend: stakeholders now seek integrated clinical–economic evidence rather than isolated clinical or financial outcomes.
Insights gained and how findings were used
The linkage revealed meaningful differences in outcomes and healthcare utilization across VeriStrat classifications, reinforcing the value of proteomic profiling alongside genomic testing in precision oncology. These findings provided evidence that real‑world performance aligns with clinical expectations and highlighted how integrated datasets can support decision‑making across payers, providers, and research partners. Following the study, Biodesix explained that additional organizations expressed interest in exploring similar linkages—an indication of growing demand for robust, multimodal RWD.
The future of linked RWD with MarketScan
MarketScan continues to expand its partnerships across lab vendors, EHR systems, and specialty datasets, enabling both turnkey and custom linkages that meet evolving research needs. As definitions of RWD broaden to include digital health sources, wearables, and device‑generated data, the need for interoperable, high‑quality, longitudinal datasets will only intensify. MarketScan’s closed claims and integrated clinical data provide the representativeness and methodological rigor required to support this next phase of real‑world evidence generation.
Want to see how MarketScan can help with your research? Connect with our team today.
Watch the complete webinar with Biodesix here.
