As Elligo Health Research’s chief data officer Michael Ibara points out, the pharmaceutical industry often is slow to embrace change, especially in areas like patient recruitment. He recently discussed with Outsourcing-Pharma some issues with traditional approaches to patient identification and recruitment and ways to fix outdated thinking.
OSP: Could you please talk about some of the ways patient ID and enrollment have progressed in recent years?
MI: The primary trends driving important changes to how patients are identified and enrolled in clinical trials are the ubiquity of digital healthcare data, and improvements in clinical informatics that allow us to take advantage of this data. As a result of regulatory incentives (in the US coming from the 21st Century Cures Act) and technological advancements, there is more computer-readable healthcare data today than ever before and it is growing at a tremendous rate. This has created a great opportunity for anyone able to apply the right skills to the data to make it usable for clinical trials.
A little-understood issue with using healthcare data is that it is not ready for use in regulated clinical trials — it must be “curated,” i.e., mapped to standard data models, standard clinical terminologies, and made HIPAA-compliant — a long series of actions needed before the data is ready for use to find patients for clinical trials. Over the last several years there is increased focus on applying the right expertise to healthcare data to make it usable in this manner.
OSP: Obviously, problems and flaws in the systems trial teams use still exist. Please talk about some of the common challenges that still occur.
MI: One of the most challenging problems to overcome in finding and enrolling patients in clinical trials is the natural hesitancy the biopharmaceutical industry has to a significant change in the patient recruitment model. Reaping the benefits of using digital healthcare data for patient identification requires us to alter our business processes and approach, and these have been institutionalized in company departments, SOPs, and expectations.
An analogous situation occurred with electricity in the 1900s; electricity was becoming ubiquitous via power plants, but it took years for electricity to increase production in traditional manufacturing. Before the value of electricity over steam could be realized, manufacturers had to “change the factory floor” so that it was better suited to a power source that was flexible and able to power individual stations versus a massive steam engine running a single massive drive shaft.
In the same way, we need to “change the factory floor” in clinical trial patient recruitment to realize the value to be found in ubiquitous healthcare data. This means we have to allow for new contracts, new ways to consider finding patients very early just from data, and new ways of paying for these services — tying a data-driven approach to finding patients to the traditional business model will limit productivity gains.
OSP: Specifically, as you point out, many studies adopt a traditional approach that relies upon sites to ID potential patients. Why does this approach prevail, and what are some of the reasons why folks still opt to go that route?
MI: As mentioned, changing an industry model is hard, and the traditional approach to finding patients is institutionalized into whole departments inside biopharmaceutical companies, in the types of contracts written, in the timelines projected, and in people’s concepts and expectations.
The industry is faced with the challenge of “changing the factory floor,” and this requires leadership and experimentation to create examples that can lead the rest of the companies in the change that’s needed.
OSP: Could you please outline some of the reasons why you think this approach is flawed?
MI: The traditional approach assumes that the scarce resource is the patient and that there is no way to know when patients will become available. Thus, the system is designed with a “wait and see” approach where you stand up a number of clinical sites and wait for patients to show up. With the ubiquity of healthcare data now available, the scarce resource is no longer the patient; rather, it’s the expertise in searching the data to find the patients. This demands an entirely new skill set and type of infrastructure setup — a data-driven approach that understands the clinical problem but also understands clinical data.
OSP: Then, please talk about ways that clinical research pros might rethink patient ID processes in order to streamline and speed up the recruitment process.
MI: First, collect healthcare data. This can be done directly from patients, clinical sites, and collective sources (e.g., lab data, genomics data).
Second, curate the data. Apply the clinical and bioinformatics expertise needed to wrangle the massive and confusing set of data from various sources into a single usable form.
Third, retool the contracting and business processes to allow for a rapid identification of patients based on data and stop restricting the speed of the process with artificial gates that are only present because they correspond to a department process set up years ago at a company.
OSP: What solutions does Elligo offer that can assist trial teams with ID and recruitment and set them up for better results?
MI: Elligo obtains healthcare data directly from patients and sites, data partners, and vendors, and we maintain this data in curated form to allow us to access a massive amount of information to find patients for clinical trials. We have the internal expertise in data wrangling, clinical and bioinformatics, and in the clinical trials and regulatory processes required to find the best matches for patients to any given clinical trial.
The 2022 Summit for Clinical Ops Executives is scheduled February 7-10, in Orlando, Florida, and online. For more information or to register, visit www.scopesummit.com.