Just by looking at the following infographic published by cbinsights, you can see a patient’s journey who is looking for clinical studies, after all forms of available treatments have failed to work.
Now let’s take a look at the structure of a clinical site and see how the AI can facilitate the process for patients. A clinical site consists of principal investigators, sub-investigator(s), clinical research coordinator(s), and depending on the size of the site a regulatory coordinator. In general, a site has so many responsibilities including but not limited to enrolling patients, conducting patient visits, submitting documents to IRB, entering data to EDC, answering queries, attending different meetings, and conducting monitoring visits. Some common obstacles for low patient recruitment in clinical sites include time, budget, human error, and access to limited data.
Time
Here is an example of a busy day in an oncology clinical site; a site with 10 different studies and 40 new and exciting patients coming in every day. For patient recruitment, the study coordinator should review all 40 patient charts. Now imagine, if the site also has patients coming in for study visits and along with that a site might have monitors coming in for different purposes (e.g. site initiation visit, interim monitoring visit, closeout visit, etc). Therefore, the research staff might not get the chance to review all patient charts. Time is critical for many clinical trials. For example, for most oncology clinical studies, if the site misses a patient in the first visit and/or in the first treatment cycle, the patient will no longer be eligible for the study.
Budget
When establishing a research department, the site needs to do a benefit-cost analysis. Hiring an employee only for patient recruitments is not often consider beneficial. Predicting patient enrollment is not easy and it requires an organized recruitment plan (check my previous article about site feasibility to learn more about recruitment plan (link here)
Human errors
In a clinical site, the study coordinator is the one reviewing patient charts, physician notes, and screening patients. The study coordinator should be a detail-oriented person. However, it is possible to miss an eligible patient due to making mistake in reviewing the chart. Although, the physicians also keep the studies in mind but, they might get overwhelmed during the day with other responsibilities.
Access to limited data
A clinical site only has access to limited info and patients. Data in Electronic Health Records (EHRs) that are reviewed by coordinators are limited to patient visits.
The AI has the ability to resolve all the above obstacles. The automated system can extract structured information such as patient demographics and clinical assessments from EHRs. In addition, these systems have the ability to identify unstructured information from clinical notes, including the patients’ clinical conditions, symptoms, treatments, and so forth. This system can have access not only to one clinical site but also to other clinics and hospitals. Finally, the system can use the extracted information and then matched it with eligibility requirements to determine a subject’s suitability for a specific clinical trial.
As Dr Kamala maddali, President of Health Collaborations stated, “Clinical Research as a care option is a right for any patient not just an option”; AI will be the enabler for this right to be accessible for everyone in a battle with any disease condition”.