Claims data is everywhere in life sciences. It’s how teams size markets, track performance, and understand prescribing behavior. But even with more data than ever before, many commercial organizations still struggle with the same problem: seeing opportunities early enough to act on them.
That tension was at the center of a recent Fierce Pharma webinar featuring Joe Zabinski, SVP of Product Management at IMO Health; Sara Zwicker, Senior Director and Brand Lead at Otsuka Pharmaceuticals; and Mayank Misra, VP of Commercial Strategy, Analytics and Operations at Neurocrine Biosciences (Soleno Therapeutics). Together, they explored where claims data remains invaluable, where it falls short, and how point-of-care intelligence can help close the patient visibility gap.
Claims data tells you what happened. Commercial teams need to know what’s happening now
The panelists were clear: claims data isn’t going away.
Its broad coverage makes it foundational for commercial analytics, market sizing, competitive assessments, and performance tracking. But claims were designed for reimbursement, not clinical insight.
As Zabinski explained, “Claims data are generated for billing… They’re not generated for clinical analysis.”
That distinction matters. By the time information appears in claims, weeks or months may have passed. Clinical context is often lost along the way. Disease severity, treatment rationale, genetic information, and provider decision-making frequently never make it into the final dataset.
According to Misra, the lag is often substantial: “The delay is roughly two and a half to three months from the time the patient saw a physician… until the time it was reimbursed.”
For commercial teams, that delay creates a challenge.
If market conditions, treatment patterns, or physician behavior are changing in real time, decisions based solely on retrospective claims data may already be behind the market.
Earlier signals can change how resources are deployed
For brand teams, access to point-of-care intelligence can change the timing of decisions.
Historically, many organizations have relied on claims-derived segments built around prescribing volume and historical behavior. Those insights remain valuable, but they’re inherently backward-looking.
As Zwicker noted, earlier clinical signals introduce a more forward-looking dimension.
“We’re not just asking what has historically been important. We’re asking what is kind of the meaningful opportunity in front of us right now.”
That shift affects everything from field deployment to messaging strategy.
More importantly, it changes when organizations engage healthcare providers. Rather than following static call plans, teams can align outreach with active decision-making windows. As Zwicker put it, the opportunity is “the difference between being present and then being useful.”
Treatment readiness isn’t a single moment
One recurring theme throughout the discussion was that patient journeys are rarely linear.
This is especially true in oncology and rare disease, where identifying patients early can have significant implications.
“Treatment readiness isn’t binary,” said Misra. Instead, he described a progression that moves from symptoms and diagnosis through referrals, testing, treatment discussions, authorization, therapy initiation, and persistence.
Each stage creates different opportunities to remove barriers and support appropriate care.
That perspective also changes how organizations think about patient finding. Rather than asking whether a patient is treatment-ready, Misra suggested asking what the next barrier is and whether we can help remove it.
Better clinical AI starts with better data
No conversation about commercialization is complete without discussing AI.
What stood out in this discussion was the panel’s emphasis on data quality over algorithmic sophistication.
Three years ago, access to advanced AI capabilities may have been a competitive differentiator. Today, the panel argued, the advantage increasingly comes from the quality, timeliness, and uniqueness of the data being analyzed.
“It’s not possible, just as kind of a rule of informatics, to get more information out of a dataset than is in the dataset, right?” said Zabinski.
Zwicker echoed that sentiment, summarizing the challenge with a familiar phrase: “Garbage in, garbage out.”
One particularly memorable observation came from Zabinski’s follow-up: today’s AI models can “make garbage look good,” increasing the risk that organizations place confidence in conclusions built on incomplete or outdated information.
Segmentation is becoming more dynamic
The discussion closed with a look at how segmentation may evolve as organizations gain visibility into near real-time patient and provider activity.
For years, segmentation has focused on relatively stable characteristics: physician type, prescribing history, patient mix, and similar attributes.
But according to Zwicker, future segmentation will be less about static labels and more about current context. That means focusing on what providers are dealing with today, what decisions they’re actively making, and where patients may be approaching treatment inflection points.
The result is a more dynamic view of opportunity, one that enables organizations to respond to changing conditions rather than simply describe them after the fact.
The bottom line
Claims data remains essential. But claims alone don’t tell the full story.
Retrospective data offers breadth and consistency, while point-of-care intelligence provides timeliness and clinical context. Organizations that can combine both are better positioned to identify emerging opportunities, reduce decision latency, and act while those opportunities are still developing.
Want to hear the full discussion? Watch the on-demand webinar, What Claims Data Misses: Finding Opportunity at the Point of Care.
