Skip to main content
The Methodologist

Real-world evidence is only as strong as the methodology behind it.

Evidence, not vibes. Strategy, not decks. Every study I design starts with a clear question and a protocol built to survive scrutiny.

0%RWD Market CAGRProjected annual growth rate through 2030 as regulatory acceptance accelerates1
0+Device Approvals with RWEFDA premarket authorizations using real-world evidence since 20162
0%Oncology RWE AdoptionOf RWE-supported label expansions are in oncology, the leading therapeutic area3

The Evidence Hierarchy

Click any level to explore — with opinionated notes from a practitioner

Data Sources — With Opinions

🏥

Administrative Claims

Imperfect, but if you understand what it captures and misses, incredibly powerful. The backbone of most RWE studies.

📋

Electronic Health Records

Rich clinical detail, messy data engineering. NLP is unlocking what structured fields can’t capture.

📊

Patient Registries

Purpose-built for research, limited by enrollment bias. Strongest for rare diseases and longitudinal follow-up.

🗣️

Patient-Reported Outcomes

The patient’s voice. Often underutilized in RWE. Combining PRO with claims data is the future.

🔗

Linked Datasets

The holy grail: claims + EHR + genomics + PRO. Hard to do, transformative when done right.

Wearables & Digital Biomarkers

The newest frontier. Continuous, objective measurement. Still figuring out analytical frameworks.

Advancing the Field

Methodologies That Move the Needle

🎯

Causal Inference in Observational Data

The bar for RWE isn't just statistical significance. It's causal clarity.

  • Target trial emulation for regulatory-grade designs
  • Instrumental variables to find natural experiments
  • Difference-in-differences for policy evaluations
🧬

Synthetic Control Arms

Historical patient data plus AI algorithms can simulate control group outcomes.

  • Cuts trial costs and timelines significantly
  • Essential for rare diseases with small populations
  • Avoids ethical issues of withholding treatment
🔐

Federated Learning for Multi-Site Evidence

Models train locally at each institution and share only analytical outputs, not patient data.

  • Privacy-preserving multi-center studies
  • HIPAA/GDPR compliant by design
  • Enables collaboration without data movement

References

  1. Yahoo Finance. Real-world data (RWD) market global report 2026: market to grow from $1.88 billion in 2025 to $2.15 billion in 2026 at a CAGR of 14.54%. Published 2026. Accessed February 2026. finance.yahoo.com
  2. Cozen O'Connor. FDA's guidance on the use of real-world evidence for medical devices. Published 2026. Accessed February 2026. cozen.com
  3. Li J, et al. Real-world evidence supporting FDA label expansions, 2022-2024. Drug Saf. 2024. Accessed February 2026. pubmed.ncbi.nlm.nih.gov
  4. US Food and Drug Administration. FDA eliminates major barrier to using real-world evidence. FDA Press Release. December 2025. Accessed February 2026. fda.gov
  5. Merative. Real-world data trends 2026: the shift to quality and AI precision. Published 2026. Accessed February 2026. merative.com

Let's Talk Evidence

Need a partner for RWE strategy, study design, or evidence generation? I'd welcome the conversation.