INDUSTRY - LIFE SCIENCES

ACCELERATING MLR approvals with GenAI

EVERSANA partnered with AWS and Modus Create to build EVERSANA ORCHESTRATE™ MLR, an AI-powered, industry-integrated platform that revolutionizes the Medical, Legal, and Regulatory (MLR) process. The collaboration focused on refining AI accuracy, scaling the platform, and implementing a multi-tenant architecture. Working with our team, EVERSANA developed a game-changing solution that automates over 90% of routine MLR tasks in the complex review process through a single platform and complies with all regulatory requirements.

OUR WORK INVOLVED

AI model optimization
Scalable AWS architecture
Multi-tenant system implementation
Process automation & knowledge sharing

IMPACT

86%

reduction in submission errors

50%

reduction in content reviews

90%+

automated tasks

35%

cost savings

Inaccuracies in marketing are costly in any industry. In the pharmaceuticals and life sciences sectors, they can be catastrophic. Over the past 20 years, pharma has shelled out billions of dollars to settle lawsuits, several of which resulted from inaccurate marketing.

That’s why the Medical, Legal, and Regulatory (MLR) process is a cornerstone of pharmaceutical content approvals. MLR reviews ensure that every claim, message, and promotional material meets strict legal and ethical standards. Beyond compliance, they help build trust with healthcare professionals and patients by delivering accurate and transparent information.

But the process is slow, tedious, and full of bottlenecks. Multiple stakeholders, complex regulations, and manual verification methods make it a roadblock for pharma companies trying to bring new treatments to market quickly.

Challenge

Improving AI accuracy

EVERSANA, a leading provider of commercial services to the life sciences industry, set out to change this. They aimed to automate MLR reviews using generative AI to speed up approvals and reduce errors.

Their team developed a proof of concept (PoC) to streamline medical claims processing. When tested, AI accuracy was between 20% and 40%. Despite the efforts of the team, the PoC was difficult to scale, and documentation was sparse, making it a challenge to build upon.

To help realize their vision, EVERSANA partnered with Modus Create to turn their PoC into a scalable solution.

Solution

Rebuilding for accuracy and scalability

Modus Create’s full-stack team, comprising a system architect, LLM engineer, product strategist, QA specialist, project manager, and front-end developers, joined EVERSANA to redevelop the PoC and take the AI accuracy beyond 80%. The approach focused on 3 phases:

Phase 1: Improving AI accuracy

The biggest roadblock was the accuracy of the PoC. If the AI couldn’t reliably match medical claims to scientific sources, the entire process would fall apart. Therefore, the team decided to improve the data quality of the model.

The existing model struggled to extract relevant information from uploaded documents, leading to incorrect results. Working with EVERSANA, our team improved data preprocessing, ensuring clean and structured inputs for the model.

Additionally, the model’s ability to link medical claims with appropriate scientific references was weak. Our team fine-tuned the model’s matching logic to provide precise and reliable results. Through continuous evaluation and iterative improvements, EVERSANA was able to see an 86% reduction in submission errors, making the PoC viable for real-world use.

Phase 2: Building for scale

A working AI model is just one part of the equation. The platform also had to handle high volumes of regulatory content across multiple clients. That’s where our work as an AWS Partner came into play, resulting in the following enhancements:

Leveraged AWS infrastructure: By using Amazon Bedrock, AWS Lambda, and Amazon DynamoDB, our team ensured the platform could process vast amounts of data quickly and securely.
Optimized system performance: The initial platform was sluggish, creating delays in approvals. Our team fine-tuned processing speeds, reducing review times from days to hours.
Implemented multi-tenancy architecture: The platform was restructured to allow multiple pharma companies to use it independently, ensuring strict data separation and customized access controls.

Adopting a multi-tenant architecture helped us ensure that the platform could scale dynamically without duplicating infrastructure costs. This approach allows resources to be shared efficiently across tenants while maintaining security and customization where needed. As demand grows, the platform can onboard new users seamlessly, optimizing both performance and cost-effectiveness.

With all these improvements in place, the PoC transformed from an internal, experimental tool to a solution ready for broader adoption: EVERSANA ORCHESTRATE™ MLR.

Phase 3: Knowledge sharing

To help improve collaboration and knowledge sharing for EVERSANA, Modus Create implemented the following best practices:

Documenting the entire platform: From architecture to workflows, we created a detailed knowledge base to streamline future improvements.
Empowering EVERSANA’s internal teams: Through hands-on training and structured knowledge transfer, our team equipped EVERSANA’s teams to manage and optimize EVERSANA ORCHESTRATE™ MLR long-term.
Streamlining collaboration: By centralizing communication on Microsoft Teams, we improved coordination between stakeholders, reducing bottlenecks and improving efficiency.

With proper documentation and training in place, EVERSANA is perfectly equipped to maintain and expand the platform without dependency on external partners.

Impact

A faster, more accurate MLR review process that scales

EVERSANA ORCHESTRATE™ MLR can now automatically identify, extract, and manage claims in minutes. Additionally, it can complete annotation and cross-referencing for creative-type documents against claims data in seconds.

EVERSANA ORCHESTRATE™ MLR is a game-changer for the life science industry. Through the power of AI and our collaborations with leading technology providers like AWS and in-house MLR experts, we’ve set a new standard for efficiency, compliance, and quality.

— Jim Lang, CEO, EVERSANA

This new platform has far-reaching implications, both in terms of efficiency and business growth. By leveraging AI, EVERSANA is redefining how pharmaceutical companies navigate the regulatory landscape, eliminating manual inefficiencies and unlocking significant cost savings.

EVERSANA ORCHESTRATE™ MLR is setting a new standard for compliance-driven content approvals. Piloted with several life sciences companies and EVERSANA’s leading agency team, EVERSANA INTOUCH, the platform has already delivered measurable benefits:

Surge in AI accuracy: EVERSANA saw an 86% reduction in MLR submission errors, making the platform reliable for production.
Saved time on content reviews: What once took weeks or months can now be completed in hours or days with the AI-powered platform, resulting in over 90% faster content updates and an overall reduction in content reviews by as much as 50%.
Massive cost savings: By automating key parts of the review process, EVERSANA ORCHESTRATE™ MLR produces 35% cost savings for most teams
Scalability without bottlenecks: The new platform can handle large volumes of MLR reviews across multiple pharma brands, ensuring efficiency at scale and significantly accelerating pharmaceutical go-to-market timelines
New business opportunities: With its multi-tenant architecture, EVERSANA can now offer the platform as a SaaS product, unlocking new revenue streams.

EVERSANA ORCHESTRATE™ MLR isn’t just about automation. It’s about unlocking a smarter, faster way to ensure regulatory compliance while opening new doors for business growth. With the right blend of AI, cloud-native architecture, and agile collaboration, EVERSANA has created a strong foundation for an AI-driven future in life sciences.

LIKED WHAT YOU READ?

You can share it with your network