Using real-world data to build a learning healthcare system

With more diverse real-world data (RWD), life sciences companies recognize the value of using the right data framework, asking the right questions and engaging expertise.

By | 3 minute read | April 29, 2020

Researcher in a lab

Life sciences teams can incorporate real-world data (RWD) into multiple stages of the drug development life cycle. The breadth and depth of RWD opens up new possibilities to identify markets and test hypotheses across populations, geographies and time. Ultimately, a strategic use of RWD can help life sciences companies see the return on investment and deliver treatments to patients faster.

The European Medicines Agency (EMA) recognizes that RWD analyses can support investigations of important questions, and has called for a learning healthcare system at the international level to help “realize the full potential of RWD.”

A learning healthcare system requires a continuous learning model grounded in three core components: 1) using the right data framework, 2) asking the right questions, and 3) engaging the right expertise.

1) Using the right data framework

Whether they work with a trusted partner or conduct the analysis themselves, life sciences teams should verify that their RWD can deliver the insights they need. RWD should be many things, including:

  • Representative – Enables large sample sizes with high-volume data sets, representing diverse populations.
  • Comprehensive – Covers the full continuum of care, including data sources such as electronic health records (EHRs) and claims databases.
  • Longitudinal – Tracks patients over long periods of time enabling identification of novel patterns which might not otherwise be prevalent
  • Compliant – Adheres to relevant standards (e.g., GDPR) and data models (OMOP).
  • Rich with clinical detail – Goes beyond patient biometrics and diagnoses to include treatments, clinical outcomes and medical history when possible.
  • Anonymized – Enables companies to share information while preserving privacy.


2) Asking the right questions

These characteristics of RWD sets are important as life science research teams seek answers to questions, such as:

How can this disease affect the patient journey?

Following the natural history of disease over time can reveal patterns on important issues, such as the burden of cystic fibrosis in subsets of populations.1

How much can this disease cost?

Research can help quantify the direct and indirect costs of a wide range of conditions, such as migraine headaches2 and cardiovascular events.3

What factors can impact medication adherence?

Understanding factors in medication delivery, such as the pharmacy4 or formulary benefits,5 can help teams address patient medication compliance.

What drives medical care choices?

Real-world data can illuminate utilization patterns and associated costs of certain treatments, such as in breast imaging.6

How can policy affect clinical guidelines, utilization and outcomes?

Government regulation and industry standards can shape healthcare services and spending, for example patterns in mental health services among individuals with eating disorders following policy changes.8


3) Engaging the right expertise

Expertise is critical to being able to succeed with the first two components. Trusted experts can help life sciences companies navigate the incredible complexities of data sets and methodologies.

A focused use of RWD throughout the life sciences life cycle can help reveal critical insights needed to bring a new drug to market. See how IBM Watson Health can help life sciences companies derive value from RWD in their clinical development, market access and commercialization strategies.

View our recent webinar: Modernize Your Value and Evidence Strategy with Global Real-World Data



  1. Hassan M, Bonafede MM, Limone BL, Hodgkins P, Sawicki GS. The burden of cystic fibrosis in the Medicaid population. Clinicoecon Outcomes Res 2018; 10: 423–431.
  2. Bonafede M, Sapra S, Shah N, Tepper S, Cappell K, Desai P. Direct and indirect healthcare resource utilization and costs among migraine patients in the United States. Headache 2018; 58(5): 700–714.
  3. Song X, Quek RG, Gandra SR, Cappell KA, Fowler R, Cong Z. Productivity loss and indirect costs associated with cardiovascular events and related clinical procedures. BMC Health Services Research 2015; 15: 245.
  4. Stokes M, Reyes C, Xia Y, Alas V, Goertz HP, Boulanger L. Impact of pharmacy channel on adherence to oral oncolytics. BMC Health Services Research 2017; 17(1): 414.
  5. Palmer L, Abouzaid S, Shi N, Fowler R, Lenhart G, Dastani D, Kim E. Impact of patient cost sharing on multiple sclerosis treatment. American Journal of Pharmacy Benefits 2012; 4 (Special Issue): SP28 –SP36.
  6. Vlahiotis A, Griffin B, Stavros AT, Margolis J. Analysis of utilization patterns and associated costs of the breast imaging and diagnostic procedures after screening mammography. Clinicoecon Outcomes Res 2018; 10: 157–167.
  7. Huskamp HA, Samples H, Hadland SE, McGinty EE, Gibson TB, Goldman HH, Busch SH, Stuart EA, Barry CL. Mental health spending and intensity of service use among individuals with diagnoses of eating disorders following federal parity. Psychiatric Services 2018; 69(2): 217–223.