How AI meets medical affairs teams’ evolving need for KPIs

This is the second in a three-part perspective on the modernization of medical affairs through applications of artificial intelligence and machine learning to drive decisions and outcomes. Part one, “AI to Align Field Engagements With Strategy in Medical Affairs,” can be found here.

The spotlight in the pharmaceutical industry is on cultivating and enhancing superior medical affairs (MA) and field teams. Their value is increasingly recognized as a critical part of successful drug launches. Because of that, it is also an area where attention is being drawn on how to evaluate and enhance team performance. 

The MA role is evolving from disease and product specialists to curators and connectors of outcomes-focused information. For these teams to have an even greater impact on the healthcare ecosystem, they must be able to measure the impact and ROI of their activities.  

What does success look like for Medical Science Liaisons (MSLs) performing well? How do you measure it? And how can you make corrections and see greater ROI? How can you track the ROI of MSLs, Key Opinion Leaders (KOLs), and Thought Leader Liaisons (TLLs) in the first place?  

Today and in the future, MA teams will be looked at to: 1) lead rapid-cycle integrated evidence generation, 2) clearly articulate clinical and economic value with partners, 3) educate and inform physicians for improved patient decision making with effective engagement, and 4) provide strategic medical direction to the organization.  

In order to do all of that well, you must be able to measure outcomes, determine value of engagements, enhance transparency between members, and align activities so that they meet strategic objectives.  

This is why MA teams need to get serious about understanding and incorporating AI into their business processes. AI can be used to measure outcomes, improve decision-making, and reveal previously hidden impact insights.  

Discerning beliefs vs. recording behavior  

If you are a leader or employee on an MA team, it won’t surprise you to learn that most of your colleagues in the industry would say that capturing and measuring the value of interactions and engagements with KOLs is one of their most intractable, enduring problems.  

These are the conversations that happen in the halls of conferences, during weekly team meetings, in performance review conversations, and in many a frustrated instant message.  

How are teams doing this now? And what would a better way look like?  

Most teams currently operate on purely qualitative beliefs and assumptions. They go about their day-to-day work lives and form individual value judgements on how important an interaction is with a particular KOL, or how impactful a presence at a specific conference is, or a media source, or online influencer. These value judgements often have more to do with an individual’s experience during an interaction as opposed to the outcomes of that interaction. 

This is not an indictment of MA team members and leaders, but rather an acknowledgment of the fact that until now it has been difficult to tie outcomes and performance to personal interactions. If you believe that a conference is valuable or a particular KOL is adding considerable value, how do you prove that other than a feeling?  

Instead of relying on individual beliefs, you can use AI technology to objectively assess behaviors and outcomes. Platforms exist today providing the ability to aggregate KOL behaviors in a single profile, making it much easier to track their individual activity: What speaking engagements have they had? What content are the publishing on social media? What research are they doing or interested in doing?  

By reviewing the behaviors of KOLs seamlessly aggregated in one operating platform, it makes it much simpler to identify how specific interactions with the MA team lead – or don’t lead – to desired behavior outcomes. 

Instead of assuming that a meeting with a KOL was highly effective, you can review that individual’s activities after the fact to see if his or her behavior actually changed. If not, it gives you an opportunity to go back to that KOL to find out more, giving MA teams the chance to clarify information or perhaps ultimately determine that a particular KOL may no longer be a good fit for the drug’s lifecycle needs. 

We know that the role of MA teams is evolving from product and disease specialists to curators and communicators of outcomes-based information. In order to meet the demands of this rapid evolution, MA teams need to be equipped with the right information at their fingertips. How are their activities and engagements really moving the needle instead of just guessing at half of the equation?   

Measuring 3rd party behaviors and inputs  

Among the changing aspects of the function of MA teams is that they must be effective scientific advocates and communicators. If that is to be a performance objective, then there needs to be more effective ways to measure the outcomes of that activity.  

Measuring that effectiveness means being able to assess how people are discussing a drug or disease online, how many prescriptions are being made, the scientific narrative, and considering if there are improvements to access or meeting unmet medical needs. 

Data analytics and machine learning algorithms make it possible to intake and process a flow of disparate external data sources, including prescription data, research journals, social media accounts, web search traffic, etc.  

It’s possible to get a clearer, more accurate picture of the value and ROI of a team’s activities by understanding to what extent the MA team member’s actions are resulting in desired, changing behavior in external metrics.  

Conclusion: Defining and measuring success  

For MA teams, measuring outcomes and the value of their activities is one of the most pressing questions they face today. Leveraging new technologies to measure outcomes and bolster performance will empower organizations to become more agile and responsive to an evolving market.