Close Icon
Article Blog November 14, 2019 4 minute read

Lessons Learned from a Life in Big Pharma

By Amit Gulwadi

Senior Vice President, Clinical Innovations, Saama Technologies

Clinical trial sponsors should focus on data, not systems, to overcome four key challenges.

For 15 years, I worked in clinical operations at two of the world’s biggest pharmaceutical companies, Celgene and Bristol-Myers Squibb, which have recently become one. During that time, I was fortunate to bring many medical treatments to market using the latest technologies available. And while the successes far outweighed the frustrations of delayed and unsuccessful trials, I knew that our industry could better leverage technology to more quickly reach patients in need.

When I started working at Saama in mid-2018, I welcomed the opportunity to apply my big pharma experience to help other ClinOps professionals improve the performance of their trials and reduce risk.

During a recent webinar, someone asked me what I would have done differently. I answered that I would have invested less in systems and more in data. The real value for sponsors is in the data that goes into their systems, not the systems themselves. You might have a great CTMS platform, but if it can’t communicate with your CROs’ systems, or if you can’t combine your CTMS data with EDC, ePRO, IxRS, eTMF, labs, etc., then you’re really not getting anywhere. It’s like the famous joke about the drunk who loses his keys as he’s stumbling out of a bar, but looks for them down the street because the light is better there.

The key for clinical trial sponsors is to find a way to take control of their data. In the webinar, I outlined four of the biggest challenges faced by clinical operations leaders:

  1. Siloed data

    Clinical research is in the business of collecting data, but data is far too difficult to access. By the time you see the data you’re looking for, it’s often out of date and there’s a ton of new data that you can’t see.
  2. The time it takes to get insights

    At every level of clinical operations, questions arise about the status of your studies. Depending on the questions, it can take up to three weeks to get answers.
  3. Resources and Excel-based tracking

    Outdated methods are inefficient, take time, and put a strain on human resources.
  4. Lack of predictive ability

    The height of frustration is finding out about an issue after it becomes a major problem.

According to a QuickPoll we took during the webinar, 50% of respondents said that siloed data was their greatest challenge, 30% gave the top spot to resources and Excel-based tracking, and the remaining 20% said their biggest challenge was a lack of predictive ability.

While none of the respondents singled out the challenge of time to insights, the idea showed up much more prominently in our second QuickPoll about the solutions attendees wanted most. Tied for first was the ability to receive context-driven insights and query data beyond dashboards and reports. A close second was the ability to detect outliers.

Data Aggregation Is the Linchpin of Good Clinical Practice

During the webinar, I explained some of the innovations we’ve been working on to address these challenges and grant these wishes. Everything hinges on technology that pulls disparate data systems and formats into a common data model—a necessary for running analytics and understanding performance in real time.

When all your data is centralized, standardized, and updated in real-time, three of the four challenges outlined above simply go away. Data silos are eliminated, so you can access whatever data you need whenever you need it. This, in turn, frees up resources and delivers insights for better decision-making.

As far as desired solutions go, data aggregation definitely speeds your ability to detect outliers. By monitoring meaningful KPIs and KRIs, alerts can tell you when predetermined thresholds are breached so you can deal with issues early on.

AI Allows ClinOps Professionals to See the Future

While data aggregation alone has been a game-changer, artificial intelligence (AI) and machine learning (ML) bring a powerful, predictive element to data analytics. AI/ML addresses the other two items on the clinical operations wish list: the ability to query data beyond dashboards and reports and the ability to get contextual insights delivered directly to you.

Data from dashboards often leads to more questions about WHY things are happening. And too often, the answers to these questions require direct contact with a programming expert, medical monitor, data manager, or other expert.

Saama’s Deep Learning Intelligent Assistant (DaLIA) is a virtual assistant that can answer questions instantly, using what I like to call Explainable AI. For example, you might want to determine the underlying factors driving a two-week delay in first site activation. Without knowing why, you’re left guessing or have to take time to figure it out. But now, in the UI, we can immediately draw your attention to the root cause so you can attack it.

DaLIA is currently based on a library of more than 3,000 questions and it’s constantly getting smarter. It can answer questions from CRAs all the way up to the Head of ClinOps, and it can push context-driven insights to the right people before questions are even asked.

If database lock is coming up in a few months, for example, our AI engine can tell you what you need to focus on, instead of making you search around for situational awareness.

At Saama, we’ve been very purposeful about the problems we’re solving. Our goal is to remove the daily frustrations and distractions of clinical trial professionals. As someone who’s been in the trenches, I can tell you that the solutions we offer today, and the ones that will be available tomorrow, will transform your clinical trial processes and lead to unprecedented results.

Saama can put you on the fast track to clinical trial process innovation.