Clinical trials tell you whether a drug works under carefully controlled conditions. Real world evidence tells you what happens when actual patients, with their comorbidities and imperfect adherence, take it in ordinary clinical practice. Regulators, payers, and drug developers all increasingly want that picture, and a substantial industry has grown up to supply it. Here is a guide to the leading real world data and evidence companies and how biotechs actually use them.
What real world evidence is for
Real world data comes from sources generated during routine care, electronic health records, insurance claims, registries, pharmacy data, and increasingly wearables and patient-generated data, and real world evidence is what you get when that data is analyzed to answer a question. Biotechs use it in several distinct ways. For regulators, it can support label expansions, provide external control arms where a randomized comparison is impractical, and satisfy post-approval commitments. For payers, it demonstrates value and real-world effectiveness, which is central to reimbursement. For development, it informs trial design, characterizes the natural history of a disease, identifies where patients actually are, and reveals unmet need. And for safety, it supports pharmacovigilance across large populations.
The major data and analytics providers
- IQVIA is the giant of the category, combining vast data assets with analytics and consulting, and its scale in both data and services makes it a default consideration for large programs.
- Flatiron Health (part of Roche) is a leading source of oncology real world data drawn from electronic health records, and it has been influential in demonstrating regulatory-grade RWE in cancer.
- Optum and Komodo Health offer large claims and healthcare data assets with analytics layered on top.
- Truveta, built from a collective of health systems, and Datavant, which focuses on connecting and linking disparate health datasets while protecting privacy, address the persistent problem that useful data sits in many silos.
- Aetion specializes in regulatory-grade real world analytics designed to meet the standards regulators and payers expect.
- Tempus combines molecular data with clinical records, particularly in oncology, and Veradigm and TriNetX provide substantial clinical data networks.
The hard part is credibility, not data
The defining challenge in this field is not obtaining data but generating evidence anyone will accept. Real world data is messy, incomplete, and collected for purposes other than research, which makes it vulnerable to bias and confounding in ways a randomized trial is not. Regulators and payers therefore scrutinize RWE closely, and the value of a provider lies substantially in their ability to produce analyses that withstand that scrutiny, through rigorous methodology, transparent study design, data quality, and appropriate handling of the biases inherent in observational data. This is why regulatory-grade methodology and demonstrated acceptance by regulators and payers matter more than the raw size of a data asset.
How to choose an RWE partner
Match the partner to the question. Consider data fit: does the provider have data that actually covers your disease, population, geography, and the variables you need? A vast dataset that lacks your specific patients is useless. Assess methodological rigor and regulatory acceptance, since evidence that regulators or payers will not accept has little value. Evaluate the specific use case, because a partner strong in oncology EHR data may be wrong for a claims-based safety study. Consider whether you need data, analytics, or both, as some providers supply data while others deliver full studies. And think about data linkage, since combining sources often produces the most complete picture and specialized providers exist precisely to enable it.
The bottom line
Real world evidence has become genuinely important to how drugs get approved, reimbursed, and understood, and the field is led by IQVIA at scale, with Flatiron strong in oncology, Optum, Komodo, Truveta, and Datavant offering major data assets and linkage, and Aetion and Tempus focused on rigorous analytics and molecular-clinical integration. The right partner depends on whether their data genuinely covers your question and whether their methods produce evidence that regulators and payers will accept. Choose for fit and credibility rather than raw data volume, because in this field, evidence that no one believes is worth nothing.
When RWE cannot replace a trial
Enthusiasm for real world evidence sometimes outruns what it can actually do, and being clear about its limits is essential to using it well. The core limitation is that real world data is observational: patients were not randomized, so the people who received a treatment may differ systematically from those who did not, in ways both measured and unmeasured. This confounding is the fundamental problem, and while sophisticated methods can adjust for measured differences, no analysis can fully account for factors that were never recorded. That is precisely why randomized trials remain the standard for establishing that a treatment causes a benefit, and why regulators are appropriately cautious about RWE used to support efficacy claims. Real world evidence is at its strongest where randomization is impractical or unethical, where the question concerns how a treatment performs across a broad, messy population rather than whether it works at all, where the effect size is large enough to be robust to confounding, and where the data genuinely captures the relevant variables. It is weakest when used to make causal claims about modest effects in settings where treatment choice was likely driven by patient characteristics the data does not capture. The practical guidance is to use RWE for the questions it answers well, natural history, treatment patterns, safety signals, effectiveness in real populations, external controls in rare diseases where a randomized comparator is genuinely infeasible, and to be honest with yourself and your regulators about the limits when you push beyond that. Used within its limits, it is genuinely powerful; used beyond them, it produces evidence that will not survive scrutiny and will waste your time and money.
A note on this snapshot
The real world data landscape is crowded and consolidating, with new entrants, partnerships, and acquisitions changing the picture regularly. The companies named here are prominent rather than exhaustive, and the right partner depends far more on whether their specific data genuinely covers your question than on their general reputation. Whatever the market does, the enduring test is the same: does this data actually contain the patients, variables, and time frames my question requires, and will the resulting evidence stand up to the scrutiny of the regulator or payer I am trying to convince?
Ask what question you are actually answering
The discipline that most improves real world evidence work is to define the question with precision before shopping for data. Vague questions produce vague analyses that persuade nobody. A precise question, one that specifies the population, the comparison, the outcome, and the time frame, tells you immediately whether a given dataset can answer it and what methodological problems you will face. Most disappointing RWE projects trace back to a question that was never sharp enough to be answerable, and the fix costs nothing but rigor at the outset.
Confirm the data covers your patients
Before engaging any real world data partner, confirm the fundamental thing: does their data actually contain patients like yours, with the variables you need, over the time frame that matters? Vendors will describe enormous datasets, and those datasets may nonetheless lack your disease, your treatments, or the outcome you care about. Ask for a feasibility assessment against your specific question before committing, since discovering that the data cannot answer your question after you have paid for it is a costly and entirely preventable failure.
For the full, maintained list of digital health and data platforms, browse the BioMed Nexus digital health and biotech SaaS directory, and see our overview of digital health and biotech SaaS platforms.
Frequently asked questions
What are the leading real world evidence companies?
IQVIA is the giant of the category, combining vast data assets with analytics. Flatiron Health (part of Roche) leads in oncology real world data, while Optum and Komodo Health offer large claims and healthcare datasets. Truveta and Datavant address data linkage across silos, Aetion specializes in regulatory-grade analytics, and Tempus combines molecular and clinical data.
What is real world evidence used for in biotech?
Biotechs use RWE to support label expansions and provide external control arms for regulators, to demonstrate value and effectiveness for payers and reimbursement, to inform trial design and characterize disease natural history during development, and to support pharmacovigilance and safety monitoring across large populations.
How do I choose a real world evidence partner?
Check data fit first: does the provider actually have data covering your disease, population, geography and needed variables? Then assess methodological rigor and demonstrated regulatory acceptance, since evidence regulators or payers will not accept has little value. Match the partner to your specific use case, and consider whether you need data, analytics, or both.
