Trial Imaging Challenges at Sites | Think Tank Thursdays

Speaker: Rajan Gopalakrishnan, MS, Director of Informatics & Information Technology

THINK TANK THURSDAYS: Welcome to Think Tank Thursdays with Yunu! Where we elevate key points from conversations with industry veterans and titans. Want to join the discussion? Scan the QR code at the end of this video to share your thoughts.

Rajan Gopalakrishnan, MS: So as part of a strategic re-orientation, you know that we started doing a couple of years ago, looking at the digital transformation of the landscape of our cancer clinical research operations, you know and applying more realistic solutions, you know, to a lot of the problems that we see here, imaging was a very big pillar, you know, to a point where we had a special group where we spoke to all the folks involved in imaging, you know, upstream downstream part of clinical research operations. And we tried to figure out what some of the key problems here were, right? So I would definitely say that, you know, the the top four recurring themes that we saw, you know, in talking to the various stakeholders, you know, one certainly operationally was, you know, turnaround times for imaging assessments, right? So, very often we have these, you know, patients who come to us maybe from, you know, Indiana or maybe from Michigan or something. And you know, they're here and we want to assess whether a clinical trial is appropriate to offer to them, and that decision is dependent on a research read. And so we see that the turnaround times were completely not compatible at all in terms of being able to make any kind of an assessment for the patient when they were here, given that they would be here for maybe half a day or, you know, whatever, even if we made them wait the whole day. So there was a big incompatibility, you know, a big gap to the point where we would have to scramble and then pick up the phone and ask somebody through an initial assessment so that we could at least make some sort of a summary judgment about - yeah, maybe the clinical trial may work may not work. That was a very big, you know, inefficiency that we saw out there and some of this actually ties in, you know, with staffing, you know, and all of those needs as well. You know, whether it is having enough radiology staff to be able to do the reads on time, whether it was, you know, the clinical research staff, you know, there was always that tension you know, about not having enough people to follow through the whole workflow and just go back to the patient and say yes, you know, you do match the criteria for this trial and we would like to offer, right? So it was all tied in pretty intimately with that, and then coming back to work through an efficiency again, you know, overall you know through the life cycle of any clinical trial, the whole aspect of around, you know, figuring out you know, what needs to happen next on for a patient, you know, given a read has been completed, you know, do we have all the information needed? What is happening with this particular patient? And, you know, but so all of these things, they tie back into the whole workflow efficiency, you know, so most of the times there were threads of emails as you've seen at many different sites, you know, so just keeping a track of like five different threads of email and sub-threads launched from their Excel spreadsheets and Word documents. So it was a highly inefficient process overall. So, you know, workflow issues were, you know, burning bright red. Then going back to what happens, you know, if all of these things have not fallen into place are data quality issues, you know, so many things communicated back and forth by notes somewhere in there, you know, somebody comes up and says, "yeah, you know, the last read was noted incorrectly, you know, it should, have been an 11mm read and not a 10mm read. And then you go back and you see it was actually 9mm in the notes. And so you're like, OK, you know, who's saying what? You know? Who's correct? So, all of the data quality and then trying to figure that out, you know, very often it would be a biostatistician or it would be a data quality person who would try to bring all of the data together and say, well, this doesn't make sense, you know, the notes say this, the image says this who's tracking what lesion? And so, data quality was a pretty big issue as well. And, you know, we spun a lot of cycles trying to get everything in a form where we could say, OK we have everything we need, now make a decision, not, you know, it could stand the rigors of an internal audit, for example. So you know, these were all like recurring themes for many different trials, which is what made us stand up an entire leg for the imaging efficiency side of things and say, well, how can we do better? How can we bring in modern generation platforms to make, you know, the life of everybody involved in a clinical trial better, you know, prevent burnout better data quality coordination, and things that we owe to do as a clinical research site.