Proper site identification is one of the major aspects in clinical trials.
Proper site identification will lead to correct site selection followed by adequate recruitment of patients required for the study. Hence, improper site identification can lead to poor patient recruitment. As per research from Tufts centre for the study of drug development, only 89% of the selected sites meet the enrolment criteria. Some sites in the list were able to meet the patient enrolment criteria only by taking double the stipulated timelines.
In clinical trials usually sites are identified by following methods. Few of them are as below:
a) Internal database: This is prepared based on the previous experience with the sites by the CRO or sponsor team.
b) Contacting SMOs having contacts with various sites of the common indication.
c) Physician dictionaries or registries.
d) Key opinion leaders and peer referrals
The above-mentioned methods have its own advantages and disadvantages. Advantages of such methods of identification include prior experience with the sites, level of confidence with the investigators, and many more. Few of the disadvantages are obsolete database systems, increase in the cost of the trials by appointing SMOs for the activities and many more.
Once the sites list has been prepared, the sites are then identified based on the key parameters involved in the protocol. Confidentiality agreement is shared with these sites to maintain the confidentiality of the protocol and the trial details. Once the site agrees by signing the confidentiality disclosure agreement, a questionnaire is shared along with the protocol synopsis.
The questionnaire consists of the questions based on the protocol requirement, number of patients that the site would be able to recruit under the investigator as per the protocol requirements, prior experience of the site and the investigator in the particular indication, the presence of mandatory equipment as per the protocol at the site, etc,. The feasibility questionnaires are further scrutinized by the CRO and sponsor to select only few sites where in site selection visits are conducted. Based on the feedback of the site selection visits, the sites are taken up for study.
Though there is stringent scrutinizing of the sites to be selected for the study, the rate of success is always doubtful. This is because each protocol has its own requirement of the patient pool. Though the indication of a particular protocol and prior utilized protocol might be similar, the inclusion and exclusion criteria, other endpoints and requirements for the study might differ. This might lead to higher level of screen failures compared to that anticipated by the investigator and the study team.
Artificial intelligence and machine learning can help to resolve this issue.
Each hospital would have a repository of patient pools with the prior history of the patients. The application of inclusion and exclusion criteria of the present study using ML to the database history or electronic health records will allow the investigator to efficiently screen and conclude more precisely the number of patients the particular site would be able to provide for the particular study requirement. This would prevent the false availability of patient pools from the site. This will also help the investigator in streamlining the appropriate patients into appropriate trials.
ML can also be applied to the internal or external databases based on the number of patients the sites were able to provide and whether the inclusion and exclusions were similar, if not what were the differences, whether the site would be able to meet the new criteria, etc., This will give the figures in precession. Also using NLP the written comments provided by any other CROs or team could be obtained as positive or negative feedback.
Using ML, the internal database system can be updated with the external registries and other database systems. This will also prevent the obsoleteness of the data as it would get updated on an ongoing basis.
Thus, artificial intelligence would help to identify which sites would be able to provide us the required patient pool for the study and which would not be able to at an earlier stage. This will help in reducing the cost involved during the study start-up phase and also would help in meeting the study timelines as per the stipulations.
If you loved this story, do join our Telegram Community.
Also, you can write for us and be one of the 500+ experts who have contributed stories at AIM. Share your nominations here.
What's Your Reaction?
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.