Blog
Overcoming 'Experience Bias' in Clinical Development Strategy
How data driven models challenge our assumptions and help us make better choices.
Rick Johnston, Ph.D., Senior Principal, Software Solutions Lead
Bruce Basson, Director, Biostatistics, IQVIA
Apr 23, 2021

When pharma leaders gather to decide what molecules they want to prioritize, it can feel a little like being stuck in a traffic jam. Everyone has their own destination in mind, and they are all in a hurry to get there, but somehow there is always gridlock.

If you’ve ever been a part of the clinical development decision-making process, you know that feeling.

Without data to identify the right development strategy, based on the company’s timeline, risk-appetite, market demands and competitive climate, decision-making involves a lot of conversations and arguments between stakeholders who often have opposing points of view.

That’s not surprising, because each stakeholder brings a different perspective to the table. Executives want their products to deliver the biggest returns. Portfolio managers have limited resources to spread across many projects. Seasoned internal clinical teams base their judgement on what worked in previous trials. And key opinion leaders (KOLs) tend to bring their own focused specialties.

All of these stakeholders bring valid opinions to the discussion, but they also all have biases based on their preferences and personal experience. And because it can take a decade or more to bring even a single drug to market, even knowledgeable experts may only be drawing on a handful of trials to base their decisions.

This ‘experience bias’ can slow down the decision-making process and, even worse, lead to the loudest or most influential stakeholder winning the argument. That can lead to the wrong decision being made, with massive implications for the company’s development costs, timelines and future success.

Bring data to the discussion

The best way to beat experience bias is by basing decisions on hard data. The wealth of global clinical information currently available around costs, timing and risks provides a powerful method to help decision makers temper their own experience with facts.

When pharma companies use data and analytic technology, they can assess each molecule based on its anticipated development costs, potential risk, projected time to deliver, and commercial value. Then they can make trade-offs among those inputs based on latest market trends, unmet medical needs, current trial results, and the commercial value of competing drugs.

This allows them to model the outcomes of each development path, still incorporating inputs from ALL stakeholders, so they can see the likely outcomes of each of their preferred plans. Teams get to acknowledge their diverse backgrounds, while still selecting the right course based on the model that best supports the company’s goals.

When pharma brings data to this process, it shortens the time to a good decision, and removes the bias and emotion, leaving only facts to guide the way. It also gives decision makers greater confidence that the molecule they chose will deliver the desired business results.

Cost, risk or revenue: which would you choose?

We’ve helped many clients eliminate uncertainty and bias through this data-driven approach, and it is transforming the way they develop new drugs.

For example, an emerging biotech company with a promising immunoproteasome reached out with questions about the relevant competitive landscape and trial benchmarks. The client wanted to assess options among five orphan indications in two therapeutic areas.

Our team used Pipeline Architect to analyze the competitive landscape and trial benchmarks across the desired indications, modeling the best and worst case scenarios for each. The solution we provided highlighted trade-offs between development time, cost, risk, and commercial projections, allowing the client to choose the indication that best aligned with their strategic goals. That helped them make the decision to drop three of the indications and prioritize the two most promising assets going forward.

In another case, we worked with a pharma company that had obtained a risk-adjusted Net Present Value (eNPV) for a molecule but wanted an independent review of the plan.

After reviewing core data sets, we determined that the eNPV assessment relied on an internal assumption that had not considered critical details of the clinical development plan. As a result, their probability of technical success and value estimates were roughly 20% higher than the data projected – contributing an extra $86M to the eNPV. Our team then lead them through proposed design changes that increased the objectively-determined probability of technical success (PTS) from 15% to 31%, and the eNPV by more than $91M.

Advances in big healthcare data and analytics technologies mean companies don’t have to rely solely on gut instinct and past experiences to guide them – nor should they. Using asset valuation models in clinical development brings precision to decision-making, allowing pharma companies to break the gridlock associated with opinion-centered meetings and reduce the impact of experience bias. This approach saves time, reduces risks, and gives decision-makers the confidence that they are making the best choice for their company, their patients, and their brand.

To learn more about IQVIA’s Pipeline Architect platform click here or contact us at PipelineArchitect@iqvia.com

Related solutions

Contact Us