The challenge

Over the past couple of years, thousands of medical insights have been entered into the company’s CRM system by medical science liaisons (MSLs) and other medical affairs roles following their interactions with healthcare professionals (HCPs), healthcare organizations (HCOs), and key opinion leaders containing data on therapeutic approaches, treatment plans and gaps, patient journey pain points, and more.  

Historically, this data has only been retrievable by the IT department, but recently, leaders within the organization have faced a growing demand from the medical affairs function requesting the ability to effectively and intelligently use these insights to enhance decision-making and engagement approaches. However, due to the unstructured nature of free-text data, extracting meaningful information from the large pool of insights has been challenging—until now. With the development of generative AI models capable of understanding and interpreting large volumes of natural language as well as generating text responses, the potential for intelligent data retrieval and analysis has become a reality.  

The solution

Intellishore helped the organization integrate the free-text medical insight data into the company’s own GPT4o model and hosted it in a separate environment to ensure people with the proper training had access to the data.  

The solution, developed focusing on value and user-centricity, is a medical GPT agent, allowing users to ask questions about the data using natural language to instantly search for and retrieve specific or consolidated insights presented as text-based answers through the chat interface. Potential use cases include:  

• Preparing for engagements with HCPs and HCOs by MLSs

• Proactively identifying common pain points and concerns regarding assets

• Evaluating and revising current treatment forms to ensure fit with patient journey 

Additionally, the solution is supported by a Power BI report, which breaks down the distribution of insights across different therapies and categories. This report guides users on what questions to ask the GPT agent and highlights potential focus areas. 

The results

The integration of the text-based data into the GPT agent allows for meaningful and fast analysis of the medical insights resulting in significant efficiency gains for employees who no longer spend several hours a week manually retrieving data for analysis.   

Surveys conducted with users of the tools shortly after the launch of the solution indicate major benefits:

“It is incredible! A much needed and very timely tool. One of the few tools whose value is obvious and does not require much explanation.” – GPT agent user. 

“The tool has fundamentally changed the way we approach medical insights analysis. We can now analyze large amounts of medical data faster and more accurately than ever before” – GPT agent user. 

Future opportunities 

The potential of implementing GPT-based agents to help retrieve and analyze qualitative data is large and goes way beyond medical affairs. Immediate opportunities include:  

• Expanding the solution to the sales organization to quickly extract meaningful information from previous HCP and HCO interactions, optimizing engagement strategies. 

• Adopting a similar solution for market access and pricing functions, where users need to search and analyze multiple pricing documents and strategies as part of their decision-making process. 

As well as many other potential use cases.