Organizing Data: Q&A with Bob Zambon

Bob Zambon
Vice president of technology
& strategic partnerships
Syneos Health

Bob Zambon, Ph.D., vice president of technology strategy & strategic partnerships at Syneos Health spoke with Pharmaceutical Executive about one of the most important topics in pharma: data. New technologies have changed the landscape, which provides both amazing opportunities and new challenges.

Pharmaceutical Executive: Data is a bit of a broad term in the life sciences industry, and data collected by different sources can often look very different. How are digital health technologies helping with this?
Bob Zambon: It's a great point, because when you look across the landscape of what's considered data formats, R&D, and even just general healthcare data, a lot of times it gets subsided into different groups. For example, there’s real world data, and then you can buy further research data, so things along those lines. Where we're looking at the total value perspective requires all of them to interplay in one way or another. Having the real-world data from someone in their healthcare journey flowed back into research and design, identifying molecules and pathways, and other new ways to have a therapy function or treat patients is the critical loop that lots of companies are trying to connect. Some are starting to do this with varying levels of success. The key thing is making the data interactable or at least able to be analyzed in a way that's meaningful. This means understanding what the limitations are of real-world data, layering in analytics, and what the capabilities are to process fields into a way that's standardized. You look at different standard approaches for things like real world data that are starting to move down that pathway. It means putting data into a format that can be universally understood and is standardized. You do this so that every time you do a research project, you're not having to re-standardize, re-baseline, and remaster all your data.

Instead, you have everything in a format that can work with each other. A lot of the work that's happened started out years ago with natural language processing. Even something as simple as OCR and translating physical documents into tech into a digital format using NLP to start the process can still be difficult. There are differences to account for, such as what some of the open text fields mean in different types of health records.

You might have data saying someone was just diagnosed with cancer, or this particular lab test came back with this particular result. This is all part of moving down that path of taking all that data and making it more accessible so you can apply additional algorithms, machine learning, and AI. Everything that people are building out on top of it is really where we're heading. In some cases, we’re already doing it.

PE: How are some of the issues related to data impacting performance-based contracting?
Zambon: As we look at real world data and the ability to leverage that data in a more actionable and direct way, it's shifting what's thought about from how we reimburse and pay for therapeutics. Before, the model was to OK a drug every time you apply therapy. You’d know things like how much certain drugs cost per pill. As we’re moving towards performance-based medicine and contracting, one of the key issues there is determining how you measure performance.

How do you measure performance in a way that the payer, patient, physician, and hospital system agree that the treatment is performing.

PE: How has the growing wearables market changed the way data is being collected?
Zambon: That's a great area and it's one that's rapidly evolving. Part of it is tied to what the actual wearables are and the data it collects. How patients interface with their healthcare is changing over time. Coming out of COVID, people are becoming much more involved in their healthcare decisions. They are getting advice online, from their physicians, and from other caregivers. Patients are starting to wear health and fitness trackers more regularly. As the hardware itself becomes more and more optimized and advanced, additional algorithms are being built into that. If you look at something like the Apple Watch, you can use that to get an EKG. We're moving from where these variables and devices were previously only able to track simple things to now where you can look at your actual heart rate, along with additional metrics like temperature, pulse, and all kinds of fun stuff.

These devices are starting to detect actual signals that can be useful from a therapeutics perspective that drives decision making that’s much more impactful than just being able to track basic measurements. Knowing your heart rate on day-to-day basis, and even potentially a minute-by-minute basis, really changes how you think about the personalization of medicine.