What Role Does Data Analytics Play in Clinical Research?

Data drives today’s world, and robust data ecosystems harness the power of information to answer our deepest scientific questions. In the clinical development industry, data ensures that studies run smoothly and effectively.

Key Highlights

We spoke with Becky Gatesman, VP, Statistical Consulting Services about the importance of statistics in data when it comes to study design and research.

Tweet
Nick Tate
Nick Tate
Copywriter

In light of World Statistics Day 2020, we’re exploring how trusted data can inform our world. To better understand how statisticians and data submissions function within the clinical development plan and study design setting, we spoke with Becky Gatesman, VP, Statistical Consulting Services.

Can you tell us about your experience and expertise with regards to data and statistics and how you came to be in your role? Can you also tell us a little about PRA’s experience and expertise in this area?

I have a Masters in Biostatistics, a PhD in Health Policy, and I've been in the industry for 20 years. I have experience in statistical methods based on my degrees and what I've done in submission work. I started out as a study level statistician in the pharmaceutical industry and I worked within that setting for 17 years until I moved over to the CRO side of things and joined PRA,. Initially, I oversaw the Statistics and Programming Group. I'm now overseeing the Statistical Consulting Group, which focuses on more high-level stats consulting. In terms of expertise,

PRA analyzes and creates output and individual studies, but PRA also helps clients write protocols and basically design their studies for approval by regulatory bodies.

Can you tell us about the importance of statistical analysis and data in the healthcare industry?

Statistical analysis is the art of summarizing data. Such analysis is important to our industry because it helps people understand whether a new treatment or new procedure is efficacious and safe by providing tangible results to scientific research questions. If we didn't have statistical rigor in our studies, people could be given drugs that don't work. More seriously, people could be given drugs that could pose a safety risk.

Before rules were imposed for marketing drug approvals that required statistical evidence of safety and effectiveness, there were some risky products being sold to the general public. Simply put, statistical analysis is a mathematical tool that helps tell a critical story on the clinical development pathway.

What role do statisticians play in clinical development and study design?

The Clinical Development Plan is a document that's put together for a product, ranging from early clinical discovery all the way to post-marketing approval by regulatory bodies. It explains what a sponsor plans to do—i.e., how many studies will be conducted, what therapeutic areas are covered, and what each study should look like. It's an all-encompassing clinical document.

The statistician is a core member of the cross-functional clinical development team. Regarding study design, you need a statistician when you design a study for a variety of reasons. They help the team understand how many patients need to be to be enrolled in a study; they help the team understand what the primary and secondary endpoints should be; and they raise different statistical concerns that must be considered when designing a study. For example, a key study concern is bias and what can be done to eliminate it. The statistician is the person who sits with the team and helps with that.

There are always statisticians at the different regulatory agencies who are part of the submission review process. It’s important that statisticians have a counterpart on the sponsor side or CRO side who can help answer the questions that come from these regulatory bodies like the FDA or EMEA.

Can you speak to the type of stats that are used in these regulatory submissions and how they can impact the submission acceptance?

All sorts of stats are used in regulatory submissions. It really depends on how the study is designed and what question it’s trying to answer. If you’re trying to get an approval of your drug or medical device, you need to make sure that whatever statistics you’re doing are outlined beforehand. That shows that you had a clear and well-constructed plan, and that you didn't just try and find a needle in a haystack after the fact.

That clear plan must be aligned with whatever statistics are accepted by regulatory agencies. Usually when you're doing this, you have a meeting with the authorities to make sure they agree with your plan. There’s all sorts of methods out there, but you need to make sure that those methods apply appropriately to your study goal and that it’s agreed upon from the beginning.

How do we ensure the quality of the data we use to craft our statistical models?

From a clinical trials perspective, there's a whole process for making sure that the data you're collecting is collected in the appropriate way. It's “cleaned”, which means a the study team works together to find and correct inaccurate records, so the data isn’t spurious and can be trusted. The data cleaning process is led by our data management colleagues, but statisticians are certainly involved in the process. Whether it’s while creating the case report forms, conducting data review, programming tables, or interpreting the output, there should be a close relationship between data managers and statisticians on a study. Close collaboration between data management and statistics ensures quality data and quality output for our clients.

How important is data transparency and sharing to reliable statistical analysis?

Data transparency is important for the validation of results and provides researchers with a shared starting point in tackling a scientific problem. Keeping in mind that data privacy is critical, it would be great if companies could share their anonymized data and be transparent in what they're doing. It goes along with the “two heads are better than one” sentiment when it comes to doing your analyses. Many companies are looking at the same things. If every one of them is analyzing the same data in a vacuum, I think less progress is made.

How can statistical analysis and data be used to help guide us through the COVID-19 pandemic?

Data is knowledge. Statistics offer a methodological way to absorb data and figure out how the pandemic is working. Many studies have been started that are hoping to put a damper on the pandemic, such as the ongoing vaccine studies and treatment studies of existing COVID victims. Behind each of those studies is data to back up what the researchers and doctors do. Data and statistical analysis are the key to understanding this pandemic and possibly future pandemics.

It’s not the quantity of data that truly makes the difference, it’s the quality. That’s why we’ve not only built one of the world’s largest repositories of data, but we’ve continued to make cutting-edge breakthroughs in how that data is captured, tracked, assembled, and translated.

We supercharge our expert human knowledge with unparalleled data insights and leverage that data to create informed solutions for developing and commercializing a drug across the product lifecycle. No matter where you are in your product lifecycle—no matter the scale or complexity of your challenge—we have the data and insights to inform more intelligent, proactive, and timely decisions to drive your product forward.

Learn more about our data insights.

You may also be interested in: