One of the purposes of technology, new technology in particular, is
to help solve previously unsolvable challenges. We’re seeing this
firsthand in the clinical research space, as increasingly robust remote
trial solutions and new capabilities via artificial intelligence (AI)
are changing how we can run studies. In the case of AI specifically,
long-practiced industry standards concerning enrollment, IT
interoperability and study management can now be addressed in ways not
possible before.
For many years, clinical trials have been planned with certain key
contingencies in mind to help ensure the collection of adequate endpoint
data. An example of this pertains to medication adherence. For many
disease states, sponsors have had to account for a certain (sometimes
high) percentage of patients that will fail to take their medication
reliably. To address predictable non-adherence, sponsors often must
enroll higher numbers of patients.
Artificial intelligence solutions bring new predictive capabilities
to the table, allowing for more optimized patient pools. AI is also
helping reduce headaches around understanding data, solving problems
with interoperability between technology platforms and allowing for
smoother-running trials.
Getting Enrollment Right
High medication adherence from your patient population has many benefits aside from the endpoint data itself. By using AI to optimize your patient pool, you can include only those patients that are most likely to show high compliance with the care plan. This means sponsors can run trials with fewer patients – they don’t need to over-enroll to account for poor performers or drop-outs. Smaller cohorts typically result in faster trials as well. All of this adds up to reduced burden on sites, as groups of highly compliant patients require less outreach and support from site teams.
The AI can help identify these types of patients. By including a one-or-two-week lead-in period, you can use technology like AiCure’s Patient Connect monitoring and engagement platform to collect adherence data from prospective research study participants. Even a short amount of time like this is enough to accurately predict dosing behaviors. Simply stick with the patients that show high adherence during this lead-in period and begin your study with a group of patients likely to quickly (and reliably) produce high quality data on your therapy’s efficacy and safety.
Improve Site Optimization and Study Management
In addition to optimizing patient enrollment, AI can help optimize site selection. By reviewing historic site data, the AI creates a profile of each site, allowing sponsors to quickly and easily see which sites meet performance criteria for a given study. With the simulation and predictive capabilities in the AI, sponsors can see different scenarios likely to take place at certain sites and make informed site selection decisions. As the trial gets underway, the AI can recommend tweaks to the site strategy based on live data so that, if changes are needed, they can be made quickly.
As patient data starts coming in and being fed into AI algorithms, new predictive models are produced and existing models are tweaked based on patient behaviors. This helps sites to know how much support and engagement patients will need, allowing them to focus efforts on the patients that need it. Further, sites can quantify patient profiles based on their behaviors, allowing study teams to better understand how patients are responding to their support. This means they can adjust support strategies quickly, as needed, so that patients get the right level of support at the right time.
For sponsors, they can group patient data by site, seeing how sites are performing and getting accurate insights into how each site is likely to perform as the study progresses. Using this information, for example, they could shift patient headcount from sites that seem likely to fall behind to sites that are performing very well.
Tackling Platform Interoperability with AI
Implementing AI doesn’t have to mean adding complexity to healthcare
IT. In the case of AiCure Patient Connect, the technology is built on a
microservices architecture, with a flexible application programming
interface (API) and scalable data ingestion capabilities. This allows it
to be quickly integrated with other technologies that are in use. For
example, it is possible to easily integrate AI with an existing
Electronic Data Capture system (EDC) and other technologies. In this
way, the AI infrastructure can be married seamlessly with the sites’
Electronic Medical Record (EMR), allowing teams to access all study data
from one dashboard.
AI-Optimized Trials
Putting advanced AI to work for your studies can streamline your
study execution, improve access to study data and help you enroll the
best possible sites and patients. Any one of these improvements will
lead to significant positive results for a trial, but combined they mean
that sponsors can run more efficient studies that reach endpoint goals
faster with the highest quality data possible.
For more information about AiCure’s industry-leading, AI-driven approach to clinical research, visit aicure.com and watch our recent webinar on this topic: https://phpstack-605076-2553551.cloudwaysapps.com/videos/now-streaming-abbvie-and-syneos-put-patient-data-to-work-with-ai
