If September’s discussion was about learning from attrition, October turns that lesson into action. The conversation now centers on how predictivity—our ability to foresee outcomes rather than merely explain them—has become the new currency of translational science. Predictivity links economics, ethics, and evidence. It is how we decide whether data are meaningful before the first patient, mouse, or model ever sees a dose. When we measure progress by foresight instead of hindsight, attrition becomes a tool for calibration rather than correction.
This month’s essays each examine a facet of that shift. Brian Berridge challenges us to confront attrition as the existential signal of inefficiency; Nick Kelley shows how aligning incentives for prediction can transform proactive health; and I close with a reflection on how translational systems can turn failure into foresight. Together they argue that predictivity isn’t just a metric—it’s a mindset for a sustainable R&D future.
Inside This Issue
Brian Berridge | Nick Kelly | Szczepan Baran |
|---|---|---|
Attrition: The Existential Threat | Finding the Missing Incentive | Landing the Plane |
How learning from failure defines sustainability in drug development | Why predictive and preventive health models need aligned incentives | When failures, feedback loops, and cultural honesty make or break science. |
Brian Berridge | Attrition: The Existential Threat
Brian opens this issue by examining why high attrition rates are not just inefficiencies but early warnings for the system itself.
He argues that the true crisis in drug development isn’t the failures we see, but the lessons we fail to capture. Attrition data should be treated as a dataset—not a disaster—where every terminated program exposes weaknesses in translation. Brian underscores that sustainability requires learning loops, not new slogans. The more we measure predictivity, the closer we come to a development model that values foresight over brute force. He calls for shared registries of failed studies and collective mechanisms to track predictive validity across models, showing that understanding why we fail is the first step to designing smarter experiments.
"Attrition isn’t failure—it’s feedback written in capital letters."
Nick Kelley | Finding the Missing Incentive
Nick builds on that argument by examining how incentives must evolve to reward prediction rather than reaction.
He describes how healthcare and pharma often profit from intervention rather than prevention, and how digital health tools are beginning to realign that equation. By quantifying long-term returns from proactive care, predictive analytics can close the gap between patient benefit and business logic. Nick shows that when payers, providers, and patients all gain from early insight, we move from managing disease to managing risk. The rise of consumer health technologies, AI-driven feedback loops, and data-driven incentives is turning prediction into an economic asset. His message is clear: the future of healthcare belongs to those who invest in foresight.
"What’s good for the patient becomes good for the payer—when prediction replaces reaction."
Szczepan Baran | Landing the Plane
In the closing essay, I connect these threads into a call for a predictive culture across the translational ecosystem.
We cannot solve attrition by force or faith; we can only solve it by foresight. Predictivity should become the universal translator across models—cellular, animal, digital, and clinical—anchored in measurable confidence rather than assumed validity. I argue that learning must be institutionalized, not improvised. Every failure should feed back into better model design, more human-relevant metrics, and transparent regulatory learning. The organizations that thrive will be those that treat knowledge retention as infrastructure, not happenstance. Predictivity, when quantified and shared, becomes the bridge between efficiency and ethics—a metric of both scientific and moral progress.
"If September was about landing the plane, October is about building one that can fly itself—with foresight as its fuel."
Innovation in drug development isn’t broken because of a lack of ideas or tools. It’s broken because too often we mistake noise for signal, hype for strategy, and platforms for solutions.
This month’s essays continue the conversation from Innovation Impact’s debut but from new angles of consequence. Brian Berridge examines attrition as the hidden architecture of inefficiency and argues that learning, not velocity, defines sustainability. Nick Kelley reframes incentives around prediction, showing how foresight can finally pay its own dividends. I close with a reflection on how predictivity can turn isolated insights into an institutional habit of foresight. Together, these perspectives push beyond diagnosis toward design: what it takes to make learning measurable and progress predictable.
Join the Conversation
This isn’t a newsletter to skim; it’s a place to interrogate assumptions and share the war stories that don’t fit into glossy decks. If a claim isn’t anchored to a patient decision, say so. Share your near misses, your “we were wrong, then we got better” moments.