Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Realising the true benefits of risk management with Brian Barnes
6 Plays9 months ago
In this episode of Clinical Data Talks, Sylvain Berthelot welcomes Brian Barnes, Director of Risk Management Strategy at BioNTech, to explore the human and strategic dimensions of risk-based quality management (RBQM).
Drawing from his rich experience across CROs, technology vendors, and now a sponsor organisation, Brian shares how his role as a “risk assessment facilitator” reflects a broader cultural shift: developing a company-wide risk-focused mindset and moving risk management upstream into study design and operational planning.
Sylvain and Brian talk about building risk frameworks from the ground up, demonstrating return on investment for RBQM and confronting resistance and misconceptions. Brian offers practical insight into why RBQM is less about the tools and more about people, collaboration, and culture.
Tune in to discover what “good” risk management looks like and how to make it stick –across teams, trials, and organisations.
Recommended
Transcript
Introduction to Clinical Data Talks
00:00:14
Speaker
Welcome to Clinical Data Talks, a podcast brought to you by CRS-Cube.
00:00:19
Speaker
I'm your host, Sylvain Bertelot.
00:00:21
Speaker
Join me and industry experts as we discuss the latest trends impacting the world of clinical data.
Who is Brian Barnes?
00:00:30
Speaker
I'm very pleased to welcome Brian Barnes to the podcast today.
00:00:36
Speaker
He has worked for a CRO, an eclinical vendor, and now a sponsor, and he's acquired a wealth of experience in risk assessment along the way.
00:00:49
Speaker
He's now the director of risk management strategy at BioNTech.
00:00:55
Speaker
So if you're interested in RBQM and risk management in general, you're in for a treat.
00:01:03
Speaker
Hi, Brian.
00:01:04
Speaker
Thank you for joining me.
00:01:05
Speaker
How are you doing?
00:01:07
Speaker
Doing very well.
00:01:08
Speaker
Thank you again for the opportunity to join.
00:01:10
Speaker
It's a pleasure.
00:01:11
Speaker
And I have high expectations.
00:01:16
Speaker
I'm sure I'm going to learn a lot from you today as you have a lot of experience in today's topic.
Transition to BioNTech and Risk Management Role
00:01:24
Speaker
What I found interesting is when we were preparing for the podcast and you described your role to me, you said that you're a risk assessment facilitator.
00:01:37
Speaker
What do you mean by that?
00:01:40
Speaker
Yeah, I think for me, coming to BioNTech, as you kind of shared in the introduction, I've come from being in the CRO space for quite a few years and then had an opportunity to develop technology.
00:01:53
Speaker
But in the CRO and technology, I didn't really have an opportunity that I felt to be able to interact and induce kind of the change that we're looking for or that
00:02:04
Speaker
industry and regulators have been striving for.
00:02:06
Speaker
So a unique opportunity came up with BioNTech where starting after Comanardi and some of our COVID-19 vaccine work, had an opportunity to redevelop our pipeline.
00:02:17
Speaker
And what was exciting is how can we build some of these risk-based strategies and risk-based thinking
00:02:23
Speaker
into early operational development.
00:02:27
Speaker
As the risk management facilitator, and my risk management role was really to start, how do we build a framework within BioNTech to support these risk-based approaches?
Developing Risk Management Framework at BioNTech
00:02:37
Speaker
The first step that I started with was defining what does a good risk management strategy look like?
00:02:43
Speaker
you know coming on the heels that we've seen recently in january iche6r3 r2 previously to that and some of the iche8 r1 really gave us good frameworks of quality by design and it's nothing new really to uh kind of the global community outside of you know clinical development it's a lot of just standard practices about good clinical kind of practices or quality by design principles but for me
00:03:12
Speaker
Bringing that to BioNTech was first kind of creating an SOP, a structure and an architecture for us to be able to start taking these risk-based approaches.
00:03:21
Speaker
So for me now as a facilitator, I've been at BioNTech for about a year and a half now, the SOP in effect for over a year.
00:03:32
Speaker
And so now as part of my role as a facilitator is how do we start implementing?
00:03:37
Speaker
And for me, I think that's the exciting piece and the component that I liked is,
00:03:41
Speaker
How do you start embedding yourself within these different cross-functional teams to facilitate these risk-based conversations?
00:03:48
Speaker
So for me, it's being kind of an active practitioner and how do we start leading our teams and our organizations to take these kind of risk-based approaches?
00:03:59
Speaker
Okay.
00:04:00
Speaker
Well, that sounds like you must have a lot on your plate.
00:04:06
Speaker
What's your goal?
00:04:07
Speaker
Because you mentioned the regulations, which obviously drive in a way the focus on risk management, but at BioNTech or maybe just you personally, what's your goal with risk management?
Goal: Participant Safety and Data Quality
00:04:23
Speaker
So I think the first piece, when we look at kind of the goals or outcomes, what are the drivers that we're looking for is how do we support participant safety and data quality oversight?
00:04:33
Speaker
Right.
00:04:33
Speaker
And that's what regulators globally have kind of given us as that anchor spot to focus on.
00:04:39
Speaker
Right.
00:04:39
Speaker
So that's the question that I always come back to and work with asking my teams and really our partners across, you know, our outsourced partner or vendor activities are.
00:04:50
Speaker
How can we focus on protecting participant safety and data quality?
00:04:53
Speaker
So those are two very simple questions that we ask.
00:04:56
Speaker
And then with that, and where I like risk based strategies is focusing on risk proportionality.
00:05:02
Speaker
And so what that means and how I kind of describe it quickly to our teams are we cannot take a one size fits all approach.
00:05:11
Speaker
And so when we do take one size fits all, we get a disproportionate focus on areas that don't bring us the greatest impact.
00:05:19
Speaker
So and we'll talk and I think this is where industry
00:05:24
Speaker
can start to lean in is driving efficiency.
00:05:27
Speaker
So risk proportionality really equals efficiency and efficiency in clinical trial development is really one of the foundational frameworks of what industry and regulators are looking for is how do we drive higher quality products to the market?
00:05:42
Speaker
really bringing them with higher quality and focus on the participant safety and data quality.
00:05:47
Speaker
Risk-based thinking and practices and risk proportionality helps us drive efficiency into that product development lifecycle.
Plan, Do, Check, Act Cycle in Risk Management
00:05:57
Speaker
That's interesting because what I hear from the industry, at least at the moment, is a lot focused on regulations, like we need to do this.
00:06:06
Speaker
But you're bringing the efficiency point into it, which I fully agree, like you would expect that from a more risk based approach.
00:06:18
Speaker
How do you think RBQM specifically can drive an increase in efficiency?
00:06:24
Speaker
Yeah.
00:06:24
Speaker
So this has kind of been my kind of goal for the last maybe 15 years is taking those risk-based approaches.
00:06:32
Speaker
So like I shared just a moment ago is focusing on what matters most, right?
00:06:37
Speaker
And so sometimes we talk about it as being like, what is the North star?
00:06:41
Speaker
So as we're talking to the teams, how do we help them and everybody cross-functioning, right?
00:06:46
Speaker
And this is kind of going back to the role of the facilitator.
00:06:50
Speaker
is as that kind of leader, how do you get everybody focus in the same direction, right?
00:06:54
Speaker
So that everybody understands what's important and then more importantly, what's not important, right?
00:06:59
Speaker
So as we're making key decisions throughout the trial conduct activities, if we're going through database logs, if we're going through interim analysis, if we're going through these different
00:07:08
Speaker
checkpoints as we're going through data cleaning and data quality oversight, we all need to be focused on the same outcome.
00:07:16
Speaker
So the cycle that I always talk to our teams about and really the structure that we're setting up in this quality by design framework is what we call a plan, do, check, act.
00:07:27
Speaker
So it's a classic quality management lifecycle.
00:07:31
Speaker
that the planning phase comes in during the risk assessment.
00:07:34
Speaker
That's where my foundational framework and passion for risk management comes in is planning in the beginning and then doing the activities.
00:07:43
Speaker
Doing would be the trial conduct, collecting the data, all of those activities that we think about in trial conduct.
00:07:50
Speaker
And then the check is the most important spot, right?
00:07:52
Speaker
So it's not like we're blindly just going to say, hey, we thought about it and we're just going to move forward in this direction, but we need to come back and check ourselves, right?
00:08:00
Speaker
And we can talk about that in more detail, but that's where metrics and data become very important is
00:08:06
Speaker
checking in the thought processes that we had.
00:08:09
Speaker
Were we right?
00:08:09
Speaker
Do we need to make adjustments?
00:08:11
Speaker
Yes or no?
00:08:12
Speaker
And that's kind of the last piece of that is making those adjustments.
00:08:16
Speaker
So based on those data and data-driven decisions, how are we adjusting our path forward?
00:08:22
Speaker
And I like to connect this back to your maybe kind of previous conversation or question about regulation, right?
00:08:29
Speaker
Everybody kind of says, oh, we have to do this.
00:08:32
Speaker
This is what the regulators are doing.
00:08:34
Speaker
What I hear from kind of reading guidance documents and conversations that we see from regulators is you're not going to get it right every time at the beginning, right?
00:08:46
Speaker
That's an unrealistic expectation.
00:08:49
Speaker
But it's the growth and the continued process improvement that is the most critical piece, right?
00:08:54
Speaker
And that's where it's a continued evolution of telling a story.
00:08:58
Speaker
And that's kind of what I, again, described to my teams is
00:09:02
Speaker
Risk management is about a story well told.
00:09:04
Speaker
And so as we go into inspections and as we get ready for audits about, it's about telling a story.
00:09:11
Speaker
You're not going to get it right at the beginning.
00:09:12
Speaker
I've never been on a trial.
00:09:13
Speaker
Maybe you have seen a trial that has gotten it right from the beginning.
00:09:17
Speaker
It just doesn't exist.
00:09:18
Speaker
And if you do have one, please share, because I'd love to do an investigation on that.
00:09:23
Speaker
But
00:09:25
Speaker
You know, it's it's how do we continue to tell that good story?
00:09:28
Speaker
And a lot of that is good documentation, good data oversight and just continuing to spin and be a little bit better every day throughout our trial conduct.
00:09:39
Speaker
Yeah, yeah, that all makes sense.
00:09:43
Speaker
There's something that frustrates me a bit about RBQM at the moment because I
Challenges in Showing ROI for Risk Management
00:09:49
Speaker
mean, I think it's a great tool and I completely agree that risk assessment is very important.
00:09:58
Speaker
However, you may not agree with that, so tell me if you don't, but I believe that what we're seeing currently is kind of a slow adoption because you've described a process that sounds quite heavy in a way.
00:10:15
Speaker
And I can imagine that, and maybe you were in that position yourself, that it's difficult to demonstrate how much return on investment the company will see, despite it, as you said, being directed at increasing efficiency.
00:10:33
Speaker
So how would you demonstrate return on investment?
00:10:38
Speaker
I think there's a couple of ways that industry has kind of thought about it before, right?
00:10:42
Speaker
And I think there's a couple of
00:10:44
Speaker
I think important technology advancements that can help us demonstrate that better.
00:10:49
Speaker
So I think think one of and I first of all, I agree with your your conversation that that change management and adoption has been less than
00:10:58
Speaker
what we would hope, but I think everything in every industry kind of takes a little bit longer than we initially thought.
00:11:04
Speaker
But I think we're on a good path.
00:11:06
Speaker
We've demonstrated through some recent work that I had supported a couple of years ago with Acro and some of the more recent work coming out of Transcelerate and some of the other industry consortia is kind of starting to demonstrate some of this ROI, which is important.
00:11:20
Speaker
but i think one of the biggest crutches that that we haven't been able to really get a clear line of sight on roi is uh we're looking for things that we have prevented happening and that's very hard to quantify right so as we start to do these early risk assessments and planning up front we're hoping trying we're trying to prevent things from happening so how do you quantify that thing that doesn't happen right in my opinion takes a little bit of a skilled
00:11:45
Speaker
practitioner to be able to properly identify and drive those efficiencies into the process.
00:11:51
Speaker
What I mean by that is, and this is coming back to the risk assessment, is being able to properly identify and clearly communicate what is important.
00:12:00
Speaker
If we have these buckets of all of the data is important,
00:12:05
Speaker
we are going to get these disproportionate use of onsite activities to oversee that.
00:12:11
Speaker
So it's really taking a very
00:12:17
Speaker
pragmatic approach to how we're identifying that data.
00:12:20
Speaker
So with that good astute lens, we can focus on the data that matters most.
00:12:25
Speaker
And then we have to, this is the second piece of risk-based quality management is driving effective mitigation strategies, right?
00:12:32
Speaker
So that's really what the output of risk management is,
00:12:36
Speaker
we need to put fit for purpose mitigation strategies into place.
00:12:40
Speaker
And those mitigation strategies are what we historically think of monitoring activities, right?
00:12:45
Speaker
Your onsite monitoring, your centralized monitoring, really the data management activities, our medical pharmacovigilance activities, all of those are risk control activities.
00:12:55
Speaker
mechanisms.
00:12:56
Speaker
And so we need to apply those in a fit for purpose way.
00:12:59
Speaker
And with that focus on what matters most, we can get those tailored to really identify and focus on those risks we are right in.
00:13:07
Speaker
And that does take a lot of inter iteration and in a very focused early mindset to drive those into budgets and drive those into the trial conduct conversations early.
00:13:18
Speaker
So I think, yes, I've seen some of those, but I've also seen where if we don't have that, that really, um, good focused identification early, we can start to get disproportionate outcomes at the end and the ROIs just aren't demonstrated.
00:13:34
Speaker
Yeah, yeah, yeah, I understand what you mean.
00:13:37
Speaker
So you mentioned technology in passing, but that's something I personally am very interested in, as you would expect, and I know you are as well.
00:13:47
Speaker
So you talked about emergence of technology over the past couple years that helps with risk-based quality management and this ROI.
00:13:59
Speaker
So do you think the technology is here now to support you in that process?
00:14:06
Speaker
Yes
Leveraging Big Data and AI in Risk Management
00:14:07
Speaker
and no.
00:14:07
Speaker
So I think the short answer is yes, but I don't think
00:14:11
Speaker
industry has fully embraced the opportunities that are there.
00:14:16
Speaker
And what I mean by that is when I kind of say, yes, I love technology, having spent a little bit of my previous career in the technology space was a great eye-opening experience for me about how to build technology and how technology development builds.
00:14:31
Speaker
On a side case, the way that we build technology is very similar to the way that we do trial conduct.
00:14:35
Speaker
And that was kind of the fun thing is when we think about these risk-based approaches and you look at lean and agile product development, it's the same principles and practices that we are trying to embed in trial conduct.
00:14:47
Speaker
So we're not introducing anything revolutionary and new with risk-based quality management and quality by design.
00:14:53
Speaker
a number of industries already apply these practices we're just trying to embrace them in in the way that we have been so rigidly doing it historically but anyway i think with technology the first kind of wave that i that that we saw that i think kind of unlocked a lot of this efficiency and optimization for us in trial conduct is
00:15:13
Speaker
kind of with big data analytics, right?
00:15:14
Speaker
I kind of think of that, your thoughts here.
00:15:17
Speaker
I've seen that maybe in the last five to 10 years, right?
00:15:20
Speaker
We've seen this emergence of big data and being able to analyze big data.
00:15:24
Speaker
And that's where I think centralized monitoring really helped support what industry was kind of doing as kind of a targeted SDV strategy beginning early, right?
00:15:33
Speaker
So just focusing on reducing the amount of onsite source data verification that happened.
00:15:39
Speaker
And that's what I said, that was maybe five, 10 years ago.
00:15:42
Speaker
centralized monitoring activities really started to grow in the space you know i think in the last five years really kind of on the heels of of covid covid was a great uh proof case and an example for us of how we can demonstrate some of the value in large data sets so great i think i think that big data analytics is available now uh but for me in the last two years i think right and you can do that is with ai and and machine learning and so that's kind of the next
00:16:11
Speaker
piece of evolution that that I'm sure everybody's heard across the globe.
00:16:16
Speaker
Right.
00:16:17
Speaker
But how do we apply that in our trial conduct?
00:16:19
Speaker
Right.
00:16:19
Speaker
So that's kind of where I see the next opportunity in terms of thinking about quality by design, thinking about risk based quality management.
00:16:28
Speaker
How do we start to embed these AI and ML technologies into the way that we do trial conduct?
00:16:34
Speaker
I just have one last quick question on AI, actually.
00:16:40
Speaker
There's talks of maybe AI helping us build technology.
00:16:47
Speaker
And I can see it because data mining, data visualizations could be done relatively easily in the future using tools that leverage AI.
00:17:00
Speaker
Is that something you think like as a sponsor, at least in your role?
00:17:06
Speaker
Yes.
00:17:06
Speaker
In the future, I don't know.
00:17:07
Speaker
I'll build my own Rbqm tool because I can do it.
00:17:14
Speaker
I think it's a very interesting opportunity, right?
00:17:16
Speaker
And I think the piece, and again, you know, for me, as I moved into the technology, I don't have a background in coding, right?
00:17:25
Speaker
I don't have a technology background.
00:17:27
Speaker
And that's why it was so interesting for me to go join that space of the industry is because I got exposure to things I'd never seen before, which was great.
00:17:36
Speaker
But for me, one thing I've always done is I've always thought about learning to code.
00:17:40
Speaker
Could I, could I,
00:17:42
Speaker
maybe get some different career opportunities, some different perspectives if I knew how to code, right?
00:17:48
Speaker
I think with AI, the piece that it does that is, you don't need to know how to code anymore, right?
00:17:53
Speaker
And I think that's what it opens up the door to more people.
00:17:56
Speaker
And I think that's what's so exciting.
00:17:57
Speaker
And that's a kind of another wrinkle and use case that we've done that I think is interesting where
00:18:03
Speaker
uh just entering data into the large language models and asking the right prompts and having a conversation correctly with the data you can actually unlock a lot of the same uh ideas that that you can with kind of a more structured um data analytics and so by no means am i saying is it going to erase you know the the great technology kind of um
00:18:28
Speaker
vendors that are out there.
00:18:29
Speaker
But I think it's a place to augment that.
00:18:33
Speaker
So I think the
00:18:34
Speaker
The short answer is, is I think AI can help us build technology because it removes maybe some of the barriers that have historically been there in terms of just skill sets.
00:18:44
Speaker
So you are able to introduce a different persona or person into there with a different maybe lens that that's thinking about things differently.
00:18:52
Speaker
And maybe that can help us unlock some of the pieces.
00:18:54
Speaker
So I will say, yes, I even use some of the prompting and
00:19:00
Speaker
soft coding to do some analytics for some of our oversight activities as we move through, right, without a formal technology piece.
00:19:09
Speaker
So I think the short answer is, is yeah, I think technology or AI can help us build better technology and also replace some of the technology vendors that may be in place today.
00:19:20
Speaker
Well, scary future for us.
00:19:25
Speaker
For you, it's good.
00:19:27
Speaker
We'll adapt.
00:19:28
Speaker
We'll adapt.
00:19:30
Speaker
Well, Brian, I just have one last question for you today, which is a bit off topic, but I love asking that question to all my guests.
Advice for Getting Started and Continuous Improvement
00:19:40
Speaker
What's the best piece of advice you've received that you consistently apply in the workspace?
00:19:48
Speaker
I think for me, the best thing I can say and the thing that I ask myself every day is how do I start?
00:19:55
Speaker
And I think just getting over the hurdle of starting is the biggest piece forward, right?
00:20:01
Speaker
And so as we look at just basic procrastination and things on tasks, and honestly, this is where I think AI has helped me is I just got to start, right?
00:20:12
Speaker
And I think getting over that.
00:20:13
Speaker
So whatever the task is, if it's,
00:20:15
Speaker
you know, in email or a risk assessment or some type of oversight or just having a difficult conversation is just starting.
00:20:22
Speaker
Right.
00:20:23
Speaker
And so it's the idea of starting and just getting 1% better every day.
00:20:28
Speaker
And so, like I said, you know, kind of earlier in the podcast is we won't
00:20:33
Speaker
you're not going to be perfect at the beginning.
00:20:35
Speaker
So removing that expectation of, oh, it has to be perfect before I start.
00:20:39
Speaker
And that's something I took from the technology space and kind of the idea of the classic MVP is like, what is the most basic version of this that I can put out and get the ball rolling?
00:20:50
Speaker
And as we get the ball rolling, we learn a lot, right?
00:20:54
Speaker
And that's kind of where I come back to the risk management principles is,
00:20:57
Speaker
let's just start and let's have a good continuous process improvement mindset and just continue to iterate right and that continuous iteration and process improvement is what's going to drive us to be better so for me it's just say let's just start what is the lowest and most simple product that i could put out today to start the conversation and we build from that
00:21:19
Speaker
that's nice I like that and you can apply it anywhere but yeah no you're right I can completely see how it can be difficult to start but then like focusing on the start rather than the big
00:21:35
Speaker
piece that you need to achieve, which most of the times feels daunting.
00:21:40
Speaker
That's a very good.
00:21:41
Speaker
Yeah.
00:21:43
Speaker
Well, amazing.
00:21:44
Speaker
Brian, thank you so much.
00:21:46
Speaker
I've learned a lot.
00:21:48
Speaker
I feel like I want to learn more.
00:21:50
Speaker
So we might do another one focused on AI.
00:21:54
Speaker
We'll see if you're up for it.
00:21:57
Speaker
Excellent.
00:21:57
Speaker
Would love to.
00:21:58
Speaker
Thanks again for the opportunity.
00:22:00
Speaker
No, thank you.
00:22:01
Speaker
And thanks everyone for watching us today or listening.
00:22:05
Speaker
It's been a pleasure and you can find more clinical data talks episodes on our website.