Most Medical Affairs teams already sit on the raw material for a real-world evidence program — thousands of field interactions, insight reports, and KOL conversations — and most of it never becomes evidence. The insights stay in free-text fields, the evidence gaps stay anecdotal, and the RWE studies that do get funded are designed two functions away from the people who heard the clinical question firsthand.
That disconnect is now a strategic liability. Regulators have formalized how real-world evidence enters approval and labeling decisions, payers are codifying how they evaluate it, and Medical Affairs is the function expected to orchestrate the response. This guide covers what RWE means for Medical Affairs specifically, how field teams operationalize it day to day, and what digital infrastructure makes the insight-to-evidence loop actually work.
Defining RWD and RWE in life sciences
Real-world data (RWD) is data relating to patient health status or the delivery of healthcare collected outside conventional clinical trials — from electronic health records, claims and billing data, patient registries, patient-reported outcomes, and digital health technologies such as wearables. Real-world evidence (RWE) is the clinical evidence about a medical product's usage, benefits, or risks derived from analyzing that data. The distinction matters: RWD is the input; RWE is the analyzed, decision-ready output.
For Medical Affairs, there is a useful third concept: real-world insights (RWI) — the qualitative understanding of why clinicians make the treatment decisions the data shows. A claims database can tell you that 30% of patients switch therapy at month four. Only a field conversation can tell you it's because of an administration burden the trial never surfaced. Strong RWE programs combine all three.
This is no longer a future-state discussion. An analysis published in 2025 in Therapeutic Innovation & Regulatory Science examined FDA labeling expansions granted between January 2022 and May 2024 and found that the proportion of approvals associated with RWE held steady at roughly a quarter each year — 23.3% in 2022, 27.7% in 2023, and 23.7% in 2024 — with oncology accounting for the largest share at 43.6%. RWE is now a routine component of how labels grow.
The payer side is moving in parallel. AMCP published real-world evidence standards in 2025 aimed at overcoming the barriers that have kept RWE out of US payer decision-making, and research in Value in Health across six HTA agencies shows reassessment processes increasingly using RWE to resolve uncertainties identified at launch — around safety, effectiveness endpoints, and treatment utilization — with oncology again leading at 55% of reassessments.
MAPS has been explicit about what this means for the function: Medical Affairs professionals are now expected to bring working knowledge of epidemiology, health economics and outcomes research, and RWE study design — not as a specialist skill, but as a baseline. Evidence generation has moved from something Medical Affairs supports to something it owns.
The practical question for a VP of Medical Affairs is no longer whether to build RWE capability. It's whether your field organization, insight infrastructure, and cross-functional governance can keep up with what regulators and payers now expect.
The role of MSLs and KOL engagement in RWE
Medical Science Liaisons sit at the exact intersection where RWE questions originate. In conversations with KOLs and treating physicians, MSLs hear the things no dataset volunteers: which patients are being excluded from a therapy in practice, where treatment sequencing diverges from guidelines, which outcomes clinicians actually weigh when choosing between agents.
The MSL Journal describes this position precisely — MSLs operate at the intersection of scientific, clinical, and market environments, identifying evidence gaps and gathering actionable real-world insights that feed HEOR and RWE study design. The problem is rarely that MSLs lack insight. It's that the insight arrives as unstructured anecdote, scattered across trip reports and email threads, with no taxonomy that lets anyone see that fourteen MSLs across three regions heard the same question last quarter.
Operationalizing the field's role in RWE comes down to four practices:
Designing MA-led RWE questions and studies
Not every evidence gap warrants a study. A useful filter for Medical Affairs leadership: Does the question affect a regulatory, payer, or guideline decision? Can existing RWD sources plausibly answer it with acceptable confounding control? Is Medical Affairs the right owner, or does it belong with HEOR or clinical development? Is there a defined audience and dissemination plan for the result? What happens if the answer is unfavorable?
Questions that pass typically cluster in territory Medical Affairs naturally owns: treatment patterns and sequencing, adherence and persistence, time to switch, real-world safety in populations the clinical trial excluded, and patient-reported outcomes — an area payers and HTA bodies are weighting more heavily in value assessments.
Rigor is non-negotiable here. RWE built on observational data carries inherent confounding and bias risk, and a poorly designed study erodes exactly the scientific credibility Medical Affairs exists to protect. Structured design frameworks such as SPACE and SPIFD exist for this reason — they force explicit decisions about fit-for-purpose data, feasibility, and validity before a protocol is written.
What good looks like
Fictional composite example. A mid-size biotech — call it Arden Bio — has a launched oncology asset and a 22-person MSL team. Within one quarter, MSLs in three regions independently tag insights describing oncologists delaying initiation in older patients with renal comorbidities, citing uncertainty the pivotal trial couldn't address because those patients were excluded.
Because the insights share a taxonomy, the pattern surfaces in the monthly triage as a single theme with 17 supporting interactions. Medical Affairs and HEOR scope a retrospective cohort study using linked EHR and claims data on effectiveness and safety in that subpopulation. Nine months later, the results are published, MSLs are trained on the data, and the evidence enters payer dossiers. The next quarter's field insights show the initiation-delay question receding — and a new question emerging about combination use, starting the cycle again.
The mechanics are unglamorous: a shared taxonomy, a recurring meeting, a decision filter, and a feedback loop. That's the playbook.
Connecting RWD sources to Medical Affairs workflows
The infrastructure question is usually framed backwards — as “which AI tools should we buy” rather than “can our systems connect what the field hears to what the data shows.” A simple way to audit your current state:
|
Layer |
What it holds |
Common failure mode |
|
RWD (EHR, claims, registries, PROs) |
What is happening in practice |
Licensed by HEOR, invisible to field medical |
|
RWI (field insights, KOL input) |
Why it is happening |
Unstructured free text, no taxonomy, no aggregation |
|
RWE (analyzed studies, publications) |
Decision-ready evidence |
Generated centrally, never routed back into scientific exchange |
Most organizations have all three layers and no connections between them. The integration point that matters most for Medical Affairs is the middle one: a Medical Affairs CRM that structures field interactions and insights in a form that can sit alongside external RWD — so the question “what are we hearing?” and the question “what does the data show?” can be answered in the same analysis. This is the core of TikaMobile's design philosophy: intelligence embedded in the field workflow itself, because an RWE program is only as good as the structured data feeding it, and structured data only exists when capture is effortless enough that field teams actually do it.
Using AI to turn RWD and field insights into RWE
AI's genuine contribution here is specific, not magical. Natural language processing can classify and cluster thousands of free-text field insights into themes no human reviewer would surface across regions. Machine learning can identify relevant cohorts in heterogeneous RWD and flag data-quality problems before they sink an analysis. Together, they compress the time between an emerging field signal and a scoped evidence question from quarters to weeks.
The dependency runs one direction: AI applied to sparse, inconsistent field data produces confident-sounding noise. Legacy CRM tools that field teams avoid using don't just create an adoption problem — they starve every downstream analytics investment of the data it needs. Fix capture first; the AI layer follows.
Governance, quality, and compliance considerations
Three guardrails keep an MA-led RWE program defensible. First, data governance: patient-level RWD carries privacy and consent obligations that vary by source and geography, and field insights containing identifiable patient details need handling rules of their own. Second, methodological transparency: pre-specified protocols, declared limitations, and resistance to the temptation to present association as causation. Third, clear cross-functional ownership: a documented split of responsibilities between Medical Affairs, HEOR, and any RWE center of excellence, so studies aren't duplicated, orphaned, or quietly commercialized.
There's also an emerging fourth responsibility. As clinicians and payers increasingly get answers from AI assistants, Medical Affairs inherits stewardship of whether accurate, current RWE is discoverable by those systems at all — a point recent industry white papers, including Veeva's work on Medical Affairs in the AI era, have begun to formalize.
How is RWE different from clinical trial data for Medical Affairs?
Clinical trial data comes from controlled conditions with strict eligibility criteria, which maximizes internal validity but limits generalizability. RWE is derived from routine clinical practice, so it captures broader populations, longer follow-up, comorbidities, adherence behavior, and real treatment sequencing — at the cost of greater confounding and bias risk. For Medical Affairs, the two are complementary: trials establish whether a therapy works; RWE establishes how it performs in the patients clinicians actually treat. Scientific exchange is strongest when MSLs can speak to both.
Where should RWE live organizationally — Medical Affairs, HEOR, or both?
In practice, both — with explicit role definitions. HEOR or a dedicated RWE function typically owns methodology, data partnerships, and analytic execution. Medical Affairs owns the question pipeline (sourced heavily from field insights), study prioritization against stakeholder needs, and dissemination through scientific exchange and publications. MAPS guidance consistently describes Medical Affairs as the orchestrator of integrated evidence plans rather than the sole executor. The failure mode to avoid is ambiguity: when ownership is implicit, studies get duplicated or evidence gaps fall between functions.
What are realistic RWE use cases for MSL teams in the next 12–24 months?
Three are achievable for most organizations now. First, structured evidence-gap capture: tagging field insights so recurring clinical questions become a prioritized input to evidence-generation planning. Second, using existing RWE — treatment patterns, persistence, real-world safety — to deepen scientific discussions with KOLs beyond the pivotal trial. Third, supporting local payer and formulary conversations with region-relevant real-world outcomes, including patient-reported outcomes, which are gaining weight in value assessments. None of these requires new study investment; all three require structured insight capture.
What are the main risks and pitfalls when Medical Affairs uses RWE?
The principal scientific risks are confounding, selection bias, and data-quality gaps inherent in observational data — mitigated by structured design frameworks, fit-for-purpose data assessment, and transparent reporting of limitations. The principal organizational risks are compliance drift (RWE shading into promotional use) and credibility loss from overinterpreting associations as causal claims. Privacy obligations around EHR-derived and patient-level data add a regulatory dimension. The consistent mitigation across all of these is rigor and transparency: pre-specified questions, declared methods, and honest framing of what the evidence can and cannot support.
How can we measure the impact of RWE-driven Medical Affairs activities?
Measure decisions influenced, not activities completed. Meaningful indicators include RWE incorporated into treatment guidelines and consensus statements, payer adoption of value dossiers containing your RWE, regulatory outcomes such as label updates supported by real-world studies, and observable shifts in treatment patterns consistent with the evidence. Leading indicators sit further upstream: the share of field insights that convert into scoped evidence questions, and cycle time from insight identification to study initiation. Tracking KPIs at both levels shows whether the insight-to-evidence engine is running, not just whether the field is busy.
The organizations that will lead on RWE over the next two years won't be the ones with the largest data licenses. They'll be the ones where a question raised in an exam-room conversation reliably becomes structured insight, then a scoped study, then evidence an MSL can put in front of the next clinician who asks. Start by auditing the weakest link in that chain — for most teams, it's the capture layer.
See how structured insight capture connects field teams to evidence generation → TikaMobile Medical Insights