mdm external service providers

RWE Part 2: Case Study – Performing Treatment Pathway Analyses Using Real World Evidence

Editor’s Note: this is Part 2 of our Real World Evidence blog series. To view Part 1, please click here.


A large pharmaceutical company sought to better understand the costs and outcomes associated with different pathways of Crohn’s Disease treatment by analyzing Real World Evidence.  The company had faced a number of challenges in performing these analyses in the past.  Multiple vendors had been unsuccessful in helping them to develop a data infrastructure that could support such complex analytics.  Prior analyses were limited to clinical trial data, without the ability to leverage valuable RWE data sources such as observational medical outcome data.  The company’s researchers missed valuable insights because their analyses were constrained by the data capacity, their analytic tools lacked a business user-friendly interface, and there were limited skills around the analytical tools.  These restrictions prevented business users from receiving the information they needed. The researchers needed more sophisticated analysis capabilities to support clinical and commercial operations and medical affairs in better understanding drug safety and efficacy over various patient cohorts.


Knowledgent worked with the company to undertake an analysis of Crohn’s Disease treatment pathways using the RWE data source anonymized claims data.  The first and most important step in the process was creating an Information Management layer that tied together the medical and demographic data points from each claim.  To do so, Knowledgent helped the client format their claims data in a Unified Patient Record (UPR).  The UPR contained thousands of data points on each patient which would normally be spread among dozens of tables in a traditional database environment. In doing so, it dramatically improved the storage and accessibility of the structured and unstructured patient data. Once built, the UPR served as a data archive, an analytics accelerator, and a patient-centric data store providing a longitudinal view of patient histories.

In addition to implementing the UPR for the client, Knowledgent provided guidance and education on Hadoop and other enabling technologies.  Knowledgent’s Informationists helped the client to merge traditional and Big Data capabilities into a single analytics infrastructure using a complementary blend of relational and leading-edge Hadoop-based infrastructures.  This provided the necessary environmental underpinnings for the requisite data manipulation and advanced analytics.  Knowledgent’s Informationists then performed several Proofs-of-Concept treatment pathways analyses and health resource utilization comparisons. These analyses were visualized through Sankey diagrams, a type of flow diagram in which the arrow width is proportional to the flow quantity.  Lastly, Knowledgent worked with the client to build self-serve analysis tools for users with limited technical backgrounds.  This will enable the researchers to be self-sufficient in conducting new and innovative explorations of the data in the future.


The client sought to identify the treatment pathways that patients with Crohn’s Disease undergo based upon relevant clinical events, including Crohn’s Disease diagnoses and surgeries, along with other medications the patient was prescribed.  By using Sankey diagrams to visually cluster the patients into different treatment pathways, the project team was able to correlate patient outcomes – such as hospitalization rates and medical costs – with the treatment pathways.  The team was also able to perform analytics on the pathways themselves, gaining an understanding of the percent of patients on each pathway and other key metrics.  The client was impressed by the intuitiveness of the visualizations along with the valuable insights they depict.

Figure 1: Crohn’s Disease Treatment Pathway Analysis

Knowledgent and the client were able to uncover new insights into drug candidates and guidance for additional clinical trials and data collection.  As a result of the project, the client now has:

  • Self-Service Analytics which give business users the opportunity to explore areas of interest across the entire product portfolio, build patient cohorts, analyze health resource utilization, inspect treatment pathways, and view disease progression
  • Broadened analytics capabilities with a toolset and infrastructure that allow more comprehensive research quickly and efficiently
  • Better insights to drive the business in the future from a much broader dataset that incorporates real-world evidence outcomes data

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.