How to Use AI to Achieve Faster Reimbursements

You’ve heard the adage that time is money. But did you realize just how much money is at stake when launching a pharma product? According to Pharmaceutic Executive, a delay in launch can cost a company an average of $15 million per drug, per day.

Delays also cut into the lifetime earning potential of the drug. In the competitive pharmaceutic industry, every day the launch is delayed is another day the competition has to try and catch up.

And that’s just the monetary cost. When you consider the impact delays have on the patients and physicians who need these products, the moral price raises the cost of delays exponentially. That’s why assisting in faster reimbursement is among the most valuable uses of AI for pharma.

But before getting into that, it’s important to understand what causes delays after a drug is approved by FDA/EMA and must be submitted to the different formularies in the US, or the different country payers if in Europe.

The goals of these payers can be different – outside the obvious one of lowering costs – depending on their specific responsibilities, and if your submission doesn’t align with them, you’re wasting time and resources. For example, showing a significant savings in hospitalization in favor of your drug would be beneficial if a payer is responsible for total healthcare costs. Both the drug cost and impact of the drug factor into that goal. However, showing data on reduced hospitalizations to a payer whose responsibility is solely around drug costs may not be the best use of that meeting.

To ensure financial success of a drug, one needs to understand and incorporate the real payer drivers (for each payer influencing the decisions) in your strategy for each stage of development and commercialization.
Changes in policy can also impact speed to market. While this isn’t an everyday occurrence, it does happen. In fact, as the industry continues to shift to more of a value-based system, it would be foolhardy not to expect policies to shift with it. Given the volume and complexity of these rules, staying on top of it isn’t always as easy as it sounds.

Artificial Intelligence gives us the means to uncover and correct the inefficiencies in drafting and submission, so pharma companies can get to market faster. The optimal approach utilizes several techniques using a combination of natural language processing and machine learning.

Using advanced data mining techniques, it is possible to gather insights based on past industry-wide experience and information on policy changes and process alterations. You likely already have much of the data needed to optimize your strategy, although it may be scattered throughout disparate internal sources. When combined with publicly available data on your competitors’ products and submissions and up-to-date information on policy change and process alterations, you can begin to uncover the factors that have the biggest impact on accelerating tie to reimbursement.

These factors include everything from the type and tone of language used to the clinical trial data and pricing levels to local market trends and everything in between. When we analyze and compare the data using machine learning algorithms and natural language processing in the context of fast vs. slow reimbursement approval, it’s possible to understand which of these are important to the decision making process and outcomes of the reimbursement authorities.

In addition to the factors selected, an AI-powered approach can also uncover previously unknown or disregarded factors that would likely lead to more effective strategies. And these insights could be used to guide future submission strategies as well. It could even enable time-to-market estimates and success probabilities for planning and budgeting purposes so you can prioritize your best chances of success.

The benefits of AI in this area aren’t limited to just gaining insights. It can also streamline the execution of those plans and procedures. Using sophisticated techniques, you can develop the submission documents utilizing both the drivers for that formulary as well as the optimal language shown in previous submissions to work best for that formulary committee.

If we know who is on the formulary committee, we can ramp this up even more by analyzing the accessible data on the individual members of the committee to identify their drivers and utilize that knowledge in how the submission documents are drafted.