RWE wrestling

Grab a chair and get to grips with the rapid evolution of real-world evidence

Real-world evidence (RWE) is entering a new phase of development, moving well beyond retrospective data analysis into an essential part of how healthcare systems learn and improve.

As life sciences organisations push towards more personalised, outcomes-focused care, RWE offers a clearer view of how treatments work outside the structured environment of clinical trials.

Once limited to administrative claims data, RWE has evolved to incorporate electronic health records, patient-reported outcomes, medical device data and more. This expansion has widened the lens through which we evaluate the performance, risks and value of interventions in routine care settings.

To achieve the full potential of a learning health system – where data from care delivery feeds directly into continuous research and improvement – several long-standing barriers must still be addressed.

These include inconsistent data quality, limited interoperability between systems and complex governance requirements around privacy and consent. The separation between research and clinical care remains a structural challenge, though one that the industry is gradually bridging.

The increased volume and complexity of real-world data (RWD) present analytical challenges. Data from different sources is rarely structured for seamless analysis, and managing the full life cycle – from capture and storage to transformation and use – requires a robust infrastructure.

Many organisations also face internal bottlenecks: the ability to interrogate data is often limited to small teams with advanced technical skills.

As data sets grow richer and more varied, traditional analytical approaches fall short. Machine learning and advanced analytics now play a central role, offering the ability to explore thousands of variables and identify patterns across diverse patient groups.

However, the hardware needed to fully realise these capabilities is still developing. Emerging areas like quantum computing may eventually provide the scale required to uncover entirely new therapeutic insights, but the foundations of data quality, consistency and governance must be in place first.

The rapid growth of digital health technologies has introduced new sources of real-time, patient-generated data. Wearables, sensors and mobile health tools offer unprecedented opportunities to track health outcomes longitudinally.

Integrating this data into the research and care ecosystem remains a work in progress, though. Much of it lacks validation, standardisation or context, requiring new methodologies to assess its fitness for purpose.

The value of these technologies lies not just in the volume of data they collect, but in how that data is interpreted and applied.

Creating meaningful, high-quality evidence from continuous data streams is still in the early stages and demands careful alignment between data science, clinical expertise and system design.

The end goal of RWE is to support more tailored care decisions rooted in evidence drawn from patients with similar profiles – not just the average trial participant.

Used well, RWE allows decision-makers to understand not only whether a treatment works, but for whom, under what conditions and over what time frame.

To get there, life sciences companies, healthcare providers and regulators must continue investing in multimodal data integration, analytical platforms and the skills to effectively utilise insights.

Solutions like synthetic data and generative AI are beginning to supplement traditional data sets: enabling simulation; modelling and faster iteration of research questions while preserving patient privacy and improving compliance.

With the right ecosystem in place, RWE can drive more timely, equitable and effective healthcare decisions. Progress may not be instant, but the direction is clear: towards a system that learns continuously; guided by data and centred around patient outcomes.

Robin Curtis is Global Marketing Lead for Health and Life Sciences. Go to sas.com