It is an age-old practice wherein the clinical research associates (CRA) travel for their allocated sites at every pre-fixed regular interval.
The CRAs need to verify every entry made by the sites and the patients. In other words, it is known as source data verification. Though it is a time-consuming method. The requirement is to complete the verification process within a limited timeline that is based on the budget of the client and the monitoring manual requirements.
This would be followed by the report preparation and follow-up letter preparation within the timelines. Again, another site allocated for the CRA would be ready for his/her visit by the time the CRA completes his letter preparations.
This continuous circle is time and money consuming. The cost of travel and the stay of CRA must be taken care of by the client. This would add to the cost of the drug indirectly. Sometimes, it might happen that the study site might have more patient count. Hence the documents that need to be verified will be huge. To compensate for the huge count, additional CRA might accompany the main CRA. Hence the cost would be increased.
Due to the tedious activity of site visits continuously along with time constraints, there are chances that the CRA might miss on the main pointers.
Based on major drawbacks faced due to traditional monitoring system, so-called Risk-based monitoring was introduced.
What is Risk-based monitoring?
According to this monitoring method, the CRA need not invest their time in complete source data verification nor the CRAs need to visit each site at regular intervals. Instead, they are asked to visit the sites which seem to be falling under the category of risk and they need to verify only the documents that are essential.
That is the complete 100% source data verification is avoided and the number of travels required to be made is reduced. This in turn would reduce the cost and time involved. Instead, the CRAs can concentrate on the critical pointers which would lead to more qualitative control of the study conduct.
The next step is to find the pre-defined triggers that lead to the identification of the risk sites.
This is possible using artificial intelligence coupled dashboards.
How exactly these dashboards work?
The data is entered into the system at various levels, that from the time a patient enters the trial till he/she is out of the trial. A few of the important data that is entered in the system are:
- The initial details of the patient such as height, weight, age, and other basic details of the patient.
- The pre-screening and screening data is entered, wherein the eligibility of the patient based on the inclusion and exclusion criteria is decided.
- The medication administered to the patient at what time and date is entered.
- The side effects experienced by the patients are recorded.
- The blood tests and other parameters tested at the regular intervals as mentioned in the protocol is also entered into the system.
These data form an important base for designing the criteria for terming a site as high risk, medium risk, and low risk. For example, if serious adverse events are over or under-reported at a single site. This triggers an alarm to the CRA for the site visit. If there are pending action items from the site to be completed for a prolonged period, this also forms a risk criterion.
Through the artificial intelligence coupled to these data, the statistical analysis of the trial data is performed during the conduct phase. This would help to reconsider, and redesign important documents involved in the study such as the communication plan, monitoring plan. This would also help to achieve the trial end at a faster rate compared to the stipulated date.
The scope for remote monitoring also increases as the source data is available in real-time in the system which can be cross verified by the offshore CRAs at regular intervals without the need for site visits.
Hence the coupling of artificial intelligence along with the database system of the clinical trial is a boon. The investigational product can be studied at a more intense level as the artificial intelligence system helps to realize the graphs which can be easily missed by humans using a traditional system.
This would help in providing a quality and cost-effective drug care system. The time taken for a product to enter the marketing system can be reduced.
Subscribe to our NewsletterGet the latest updates and relevant offers by sharing your email.
Join Our Telegram Group. Be part of an engaging online community. Join Here.
I am a passionate python coder and data scientist. I love to explore the capabilities of artificial intelligence in the field of the health care industry.