How One Bio Company Tackled Its Data Visualization Challenge
By Ed Miseta, Chief Editor, Clinical Leader
Data visualization is a problem for many life science companies. Being able to visualize reliable data in real-time, and make smarter decisions faster, is vital to delivering speed and quality across clinical trials.
While many companies struggle with quality uncertainty and surfacing relevant data trends, those that specialize in rare diseases have a more unique challenge. With rare diseases, trials are open label (patients and researchers both know which drug is being administered) and non-comparative. By design, personnel have access to data in an unblended way. That provides the opportunity to see actual data as it emerges, enabling personnel to understand the data in a way that would not be possible in a blended trial.
I spoke with a bio company that specializes in rare diseases. For clinical trials, it has a CRO handle traditional data management using a validated data management platform. From that EDC platform the data can be dumped into an Excel spreadsheet and manipulated, often by data modules.
“Let’s say you’re looking at an adverse event module,” says the company’s VP of clinical operations. “You can do a data dump from your adverse events, but then if you want to do demography, you have to perform another data dump in Excel, and you have to merge those two together. It’s not a simple process.”
They had the option to go to the CRO who would clean and code the data requested. If clinical personnel wanted to review lab data to see if there were any safety trends emerging, a statistician at the CRO would generate the same level of graph that might be used as part of a final study analysis. It might take two or three weeks to get the information, and cost up to $2,500. But during the course of a trial, the study team needed continuous “quick and dirty” monitors around particular trends. The information offered by the CRO took too long to produce, was too costly, and often was outdated by the time it was available.
This meant the data presented problems for those needing insight. Instead of looking at individual patients, personnel had to look at trends in the overall data. What was required was the need to be able to visualize the data, see what’s happening, grasp it, and be able to integrate it. For that, a solution was needed that could provide on-demand reporting and would also not require a lot of in-house expertise in the area of programming or other special technical skills. The solution would have to be flexible enough to allow physicians and clinical operations personnel to work with the vendor to develop a suite of custom reports, but also to be able to build their own drag-and-drop methods.
Aggregate Deviations Across Trials
When I asked the VP how the company addressed these issues, he said they implemented a continuous quality solution from Comprehend Clinical, which sits on top of its EDC system and enables data insights through visualizations. By selecting relevant data and presenting it in a table or graph, physicians and clinicians have the ability to focus on what matters, analyzing the data real-time as it is being accumulated.
The solution allows the research team to have oversight of their trials and monitor adverse event trends or other important factors. The company’s access and ability to monitor the data is now more on-demand and requires less programming. Some of the visualization solutions can actually generate output for interim data monitoring committee meetings to evaluate. By having access to efficacy data, the company is also able to go public with early trial results much sooner.
The company executive also noted the solution allowed them to aggregate protocol deviations and violations across all trials by using the solution’s integrator feature. “If I wanted to know how many protocol deviations have occurred, I can look at that across the portfolio or by individual study. This has the potential to drive change, as we can get out to sites and reeducate personnel. It also allows us to focus on a particular area of the study.”
As a final added benefit, the continuous quality solution forced people who generally did not look at data to review it more frequently, which allowed the company to spot more outliers. For example, clinicians might be looking at patient liver enzyme levels on one chart. If one patient had a large spike in these levels, Comprehend’s solution makes it easy to actively investigate if it was an aberration or if a decimal point was entered incorrectly. If those same clinicians were reviewing data coming out of the EDC system, a safety or quality issue might not get caught.
At this time, the bio company has only touched the tip of the iceberg, since data management expertise still does not exist in-house. Operations personnel are starting to build that expertise and are in the process of bringing in a data manager. Once that happens, it will be able to optimize usage of the program and get even more value out of it.