Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
The challenges and promises of the clinical data lifecycle with Doug Bain
5 Plays11 months ago
For the first episode of Clinical Data Talks, host Sylvain Berthelot welcomes industry expert Doug Bain to explore the intricate journey of clinical data - from its initial collection to final submission. With a background spanning software development and leadership roles at eClinical vendors and CRO, Doug brings a wealth of experience and sharp industry insights.
Together, they discuss the evolving challenges of managing clinical data efficiently, the impact of new technologies on data workflows, and the importance of standardization in the world of clinical data. Doug also shares his perspective on how the industry can address persistent issues like data fragmentation and interoperability while embracing innovations that enhance data integrity and usability.
Tune in to learn more about the clinical data life cycle and what’s shaping its future.
Clinical Data Talks is a podcast brought to you by CRScube, a leader in eClinical technology solutions.
Recommended
Transcript
Introduction to Clinical Data Talks
00:00:13
Speaker
Welcome to Clinical Data Talks, a podcast brought to you by CRS-Cube.
00:00:19
Speaker
I'm your host, Sylvain Berthelot.
00:00:21
Speaker
Join me and industry experts as we discuss the latest trends impacting the world of clinical data.
00:00:30
Speaker
A lot of you will know my guest today for his sharp observations of our industry.
00:00:40
Speaker
He started as a software developer and has since contributed to the success of a lot of eclinical vendors.
00:00:50
Speaker
He's one of my favorite LinkedIn voices and he's also known for wearing a kilt from time to time.
00:00:59
Speaker
It's no one else but Doug Bain.
Meet Doug Bain: Industry Insights and Curiosities
00:01:02
Speaker
Hi Doug, how are you doing?
00:01:04
Speaker
I have to disappoint you, I'm not wearing a kilt but I'm not even going to demonstrate.
00:01:09
Speaker
You can get the top half.
00:01:11
Speaker
I'm very glad to be on this podcast, so thank you for the invite.
00:01:14
Speaker
Well, thank you for joining me.
00:01:17
Speaker
Today we're talking about the clinical data lifecycle, and I can't think of anyone better to discuss that with, so that's great.
00:01:26
Speaker
And when you came to me with this topic, I thought, yes, that's a great topic for the podcast.
00:01:33
Speaker
And then I thought, oh, wait, what does he mean by that?
00:01:37
Speaker
First question, could you explain to our audience what you mean by the clinical data life cycle?
Understanding the Clinical Data Lifecycle
00:01:45
Speaker
I mean, I think it's something that I have sort of considered really going right way back to when I first got involved in clinical research.
00:01:54
Speaker
And I wouldn't even go, I won't even share how far back that is.
00:01:58
Speaker
But the life cycle of clinical trial data is
00:02:03
Speaker
um the steps that data needs to go to from when it's originally identified wherever its source might be all the way through to the other end where it's then used as part of an analysis to go into a clinical study report okay so that life cycle can involve edit checking it could be source data verification it could be even transposing it for an EHR basically it's going through a number of steps
00:02:31
Speaker
before you can say, yep, it's all there.
00:02:35
Speaker
And I'm interested in it because I believe there are significant breakpoints or lag points that when combined slow the whole process down.
00:02:47
Speaker
And I've been considering where are these break or lag points and what can we do to try to compress them so that you can make decisions early.
00:02:56
Speaker
You may not need to make decisions early for different reasons.
00:02:59
Speaker
but I'd like to enable the choice of being able to make these decisions, not based on the fact that data is sitting somewhere in a silo for a month or two months or more.
00:03:09
Speaker
And when you talk about decisions, you mean decisions on the clinical trial?
00:03:14
Speaker
Is that right?
00:03:16
Speaker
Principally, but I think you could also apply this in principles to some trial management, you know, clinical operations.
00:03:23
Speaker
You know, if you're trying to manage the trial, make decisions in a trial related to things like recruitment or such things, the fact you've got to wait five weeks to get a true picture of, for example, data that might come from monitoring report,
00:03:39
Speaker
that seems like five weeks that you could be doing something right.
00:03:43
Speaker
Why do we need to wait five weeks?
00:03:44
Speaker
So, so certainly the clinical data, I think the ability to be able to make decisions.
00:03:49
Speaker
Um, and I think also
00:03:52
Speaker
you know, the power of these decisions are dependent on the cleanliness of the data.
00:03:57
Speaker
If you can in principle look at data, you know, as soon as it's saved into an EDC product, in principle, you could look at that data for certain purposes.
00:04:08
Speaker
But the value, the power you get from looking at that data is going to be lessened because it's maybe not been edit checked or submitted or STV or whatever.
00:04:17
Speaker
So the power should increase.
00:04:19
Speaker
And I think,
00:04:21
Speaker
I mean, I'm sure we'll talk about AI in this podcast more than once, but let me start.
00:04:26
Speaker
The power of the ability for AI to make decisions with an appropriate level of confidence is dependent on some of these factors, knowing the label of cleanliness of that data and go, right,
00:04:39
Speaker
I can confidently say that you can make that trial adaption or something like that at this point in time.
00:04:47
Speaker
If you don't have that meta information or that state of data as early as possible, you've got lag.
00:04:52
Speaker
You eventually don't need it.
00:04:56
Speaker
Yes.
00:04:57
Speaker
Yeah.
00:04:57
Speaker
And I'm sure that's something study teams really want that would release pressure on data management as well, I can imagine.
00:05:07
Speaker
So at your end, in your experience, well, first of all, do you have experience setting up eClinical ecosystems and how did that support trial and data management teams?
Building eClinical Ecosystems
00:05:25
Speaker
So, I have and this is something that I've had more experience in the last five years working at a CRO.
00:05:35
Speaker
Prior to that, I was far more focused on
00:05:38
Speaker
How can we create the best software, either trial management or EDC most typically, or more of a DCT platform as well?
00:05:48
Speaker
So I've been focusing on that, but in the CRO world, you're saying, right, we need to create an environment for running this clinical trial.
00:05:56
Speaker
And that environment should be as solid and as reliable as you can do.
00:06:01
Speaker
At the CRO, you don't want to be adding risk into the possibility of your clean execution of that trial.
00:06:08
Speaker
So you don't want to throw risky software, risky situations and so on.
00:06:13
Speaker
So...
00:06:15
Speaker
So I have been responsible for defining the map of systems that are applicable and to ensure that that map is as connected as possible.
00:06:28
Speaker
And when I say connected, I mean you've got your EDC system speaking to your CTMS system or your CTMS speaking to, you know, et cetera, et cetera.
00:06:39
Speaker
And I feel very strongly, I blogged about this in the past, and the level of connectivity you achieve with your clinical research platform.
00:06:49
Speaker
I call it platform, but what I'm saying, when I call it platform, I'm not necessarily saying one piece of software.
00:06:56
Speaker
I do like single instance software, but you might be seeing a number of pieces to that puzzle, but make sure they're connected.
00:07:03
Speaker
And it's not just connected from a technical perspective, it's also connected from the people and the process perspective.
00:07:12
Speaker
I've seen many times where
00:07:15
Speaker
you have a stack of SOPs by department and their stack is very much geared towards this is what my department does.
00:07:22
Speaker
They don't have a strong connectivity and reflection of activities of other departments, which to me, come back to my earlier point, is putting lag into the flow because you're actually saying, I'm not going to look at that data until this big marker has occurred, which could add lag, which maybe isn't required.
00:07:43
Speaker
Yeah.
00:07:44
Speaker
So is it something that in your experience or with what you've done, teams are open to like breaking silos?
00:07:57
Speaker
To be honest, not often, I'm afraid, because I think, I mean, there is a benefit in defining silos.
00:08:06
Speaker
If you are an organization and you want to be able to scale up an organization, it's easier to scale it up by silo in some regards.
00:08:19
Speaker
I think there's alternative methods.
00:08:20
Speaker
You know, I'm a big fan of agile.
00:08:23
Speaker
I come from a software world and agile,
00:08:26
Speaker
um it's it's actually maybe not a great word you know i remember saying the word agile to an scdm meeting um last year and someone gave me a very stony face i'm going okay that's not a good word to use um but my my feeling is that you can pull together a team from a resource pool and you can say right guys we're working on this study okay and try to spin around this problem of the either the
00:08:55
Speaker
a definition of the study or the building of the study or the execution of the study, but spin through these cycles far faster than if you were to do it with a gate process where somebody finishes a department or a function finishes what they're doing and then passing that over to the next function.
00:09:12
Speaker
They do their stuff and pass the next thing.
00:09:14
Speaker
I think that slows things down.
00:09:17
Speaker
And it also doesn't lend itself towards an iteration of, hold on, that's not quite right.
00:09:22
Speaker
Maybe we should change that.
00:09:23
Speaker
So in the software world, we've sort of learned if you want to get to the best product, you should rapidly iterate through this because you can see it, you're in the moment, you're working on it.
00:09:36
Speaker
And if you don't, your focus isn't on it.
00:09:39
Speaker
You may be getting a document.
00:09:41
Speaker
Like for example, a protocol, you could send a protocol off to data management.
00:09:45
Speaker
They're looking through it.
00:09:46
Speaker
Yeah, yeah, yeah, it looks okay to me.
00:09:48
Speaker
To me, the engagement is not what it needs to be when you're reviewing a protocol.
00:09:53
Speaker
It can be very hard, but gather around a sort of meeting room or virtual meeting room where you're sort of walking through it, working with it.
00:10:01
Speaker
To me, you're focused, you're engaged, you're involved in it.
00:10:03
Speaker
And I think that creates better outcomes.
00:10:06
Speaker
Yeah, it's interesting because I've worked with a cross-functional team at one of the top farmers.
Navigating Cross-Functional Team Dynamics
00:10:16
Speaker
I was on the vendor side, but their structure was very well set in a way because it was a true cross-functional team.
00:10:24
Speaker
But I could tell that they all had their own priorities.
00:10:27
Speaker
And although the team worked very well together,
00:10:31
Speaker
I'm not sure they achieved what they wanted to because of those diverse priorities.
00:10:42
Speaker
But everyone is interested in clinical data and has one objective, which is linked to clinical data.
00:10:50
Speaker
So I think it's a right approach to focus on clinical data to bring a cross-functional team together.
00:10:57
Speaker
Yeah, I mean, I see where you're coming from and I can understand that.
00:11:00
Speaker
You know, if you're, let's say you're in Biostats, you're invited to this meeting, right?
00:11:06
Speaker
There could be a tendency for you to defend your Biostats position, right?
00:11:11
Speaker
Fighting for Biostats because these guys in data management or wherever, they'll be pushing it off this way.
00:11:17
Speaker
And that, I think you need to work on the culture of this where it's almost like you need to have a leader
00:11:24
Speaker
And a leader is not in biostats, they're not in project management.
00:11:27
Speaker
You need to say, no, this is the clinical trial leader, right?
00:11:31
Speaker
And they're helping to coordinate this team, which is one team, work.
00:11:36
Speaker
And by the way, I'm a big fan of this team, including the sponsor or including the CRO.
00:11:41
Speaker
To me, they're all around one table.
00:11:43
Speaker
They've all got role to play.
00:11:44
Speaker
There's no us to them, all around one table and they're all working to be constructive.
00:11:49
Speaker
So I think there's a cultural change, but you do need to maybe look at the organization structure and say, actually, there's a clinical trial leader, all the sets above these departments.
00:12:00
Speaker
They are empowered to say, right, we can make decisions.
00:12:03
Speaker
I mean, I think you see that more in the biotech world where you get smaller clients, they're looking for agile study delivery and they want to brainstorm, they want to create the best solution and they don't want to be told, no, this is the way we do it and sorry, you need to just accept it.
00:12:21
Speaker
This is the way we work.
00:12:23
Speaker
They're looking for a more agile team to be able to say, yeah, we can do this.
00:12:28
Speaker
Yeah, and sometimes you see solutions that don't solve any problem.
00:12:34
Speaker
Yeah, I think there's almost like you've got all the technology and then you validate the technology, implement it and you go, right, that's good.
00:12:45
Speaker
I did actually an analysis
00:12:49
Speaker
You know, KCR, I thought I'm going to sit down with some of the guys and see what you're doing.
00:12:55
Speaker
Can I see what you're doing day to day?
00:12:58
Speaker
Right?
00:12:59
Speaker
Because I want to check.
00:13:00
Speaker
I really want to understand.
00:13:02
Speaker
And this is a few years ago now, but I used this as an exercise to validate that the systems we were using were functioning, were operating the way we expected and wanted them to work.
00:13:16
Speaker
yeah um and i sat down with a cra uh and she said right um and this is my word document and and i opened this word document and i fill this word document out and i said hold on stop stop why are you filling out she said well um uh we're using i wouldn't say which system it was but we're using an edc system
00:13:39
Speaker
and the sponsor wants us to use the CDT system, but it's not been set up in a certain way, so I can't use it for data review because that wasn't enabled.
00:13:48
Speaker
Therefore, what I do is I copy paste the data into my Word document, I review it here and I interchange Q&A with the site that way.
00:13:59
Speaker
And I going, did anybody not speak to you about this?
00:14:02
Speaker
And go, you know, could we not go back to the sponsor and say, maybe you could switch on that data review flag and maybe we wouldn't need that.
00:14:09
Speaker
No, no, we've got no control over that.
00:14:12
Speaker
So that, I mean, a small example, but you know, there's many, many other cases that was Word document, but there's also Excel, you know, Excel is used everywhere.
00:14:26
Speaker
A lot of the CIOs, CTOs, they don't even want to look, they don't want to lift the cover and see where it's being used, but it's been used far too extensively.
00:14:37
Speaker
And I don't blame the users in any way.
00:14:40
Speaker
They're solving the problems that they have and they've picked Excel the tools that they can work with to solve these problems.
00:14:48
Speaker
And I think as technologists, we need to say, well, hold on a minute.
00:14:52
Speaker
If our EDC tool is not solving that problem, we need to fix the EDC tool or configure it so it does fix things.
00:14:59
Speaker
Yeah, yeah, yeah.
00:15:02
Speaker
It's interesting.
00:15:03
Speaker
I love doing that, actually, like talking to users.
00:15:08
Speaker
I used to do it in previous roles and then help shape the product.
00:15:15
Speaker
I think that's the most exciting things of technology development.
00:15:22
Speaker
But then when you think about the person who's using Excel, for example, they might be thinking, oh, yeah, but first of all, who do I talk to?
00:15:34
Speaker
And then how long is it going to take?
00:15:37
Speaker
I was listening to an audio book this morning about it's something in marketing, but which
00:15:45
Speaker
like a concept I really like like doing tests trying to do your your development life cycle so short that it becomes very easy to to test a new functionality or test a solution to a problem and I think sometimes when I think about our industry I feel like we're
00:16:09
Speaker
we're not challenging our timelines enough.
00:16:14
Speaker
I'm not talking about clinical trial development, but like the other day, it was at Scope actually, I heard someone talking about the regulations.
00:16:30
Speaker
I can't remember which one they were talking about, but
00:16:33
Speaker
Their comment was, oh, this will need to evolve because it's not optimal.
00:16:42
Speaker
But I know it's going to take 10 years to evolve, so we're going to have to live with it.
00:16:47
Speaker
I was shocked.
00:16:48
Speaker
I thought, well, if the next iteration is 10 years down the line, then that's not evolution.
00:16:55
Speaker
It's just very, very slow progress.
00:17:02
Speaker
But I would agree, you know, it is very slow and I think the regulators, they've got a challenging job to fulfill.
00:17:12
Speaker
They need to reflect current thinking, but you've almost got a chicken and egg situation.
00:17:18
Speaker
The vendors are going, no, we can't do that because that doesn't meet the current regulations.
00:17:23
Speaker
So they try to push the boundary, but it's very difficult, right?
00:17:27
Speaker
So we've seen, to me, an example of that is this,
00:17:30
Speaker
the interpretation of e-source, you know, I think some of the boundaries need to be pushed harder.
00:17:37
Speaker
You know, I was doing electronic source clinical trials directly into an EDC system in 2001, 2002.
00:17:44
Speaker
Right.
00:17:47
Speaker
And yet, you know, 20 plus years later, people are still telling me you can't do that.
00:17:53
Speaker
yeah right yeah um and that worked you know to me that was working really well i was in that case it was a phase one study we did 49 studies in that phase one setting
00:18:06
Speaker
and we wanted to capture data at source, but there was a perception that, oh no, you'd had to keep that data at the site.
00:18:14
Speaker
And there's still the perception of the fact that that data has to be at the site or under the control of the investigator or the site in a separate database.
00:18:26
Speaker
But I just completely disagree with that one.
00:18:31
Speaker
Software technologies are creating these crazy separation systems.
00:18:36
Speaker
You've got one database for the investigator and a separate database sitting on the same server or hardware to keep it separate.
00:18:46
Speaker
This is what happened.
00:18:47
Speaker
A number of vendors are doing that and they're saying, no, no, no, we need to keep it separate so they've got control.
00:18:53
Speaker
It makes no sense, you know, because effectively you're just using bits and bytes to set permissions to gain access, whether it's in the same physical database or in two separate databases, the same, you know, mechanism that achieves it.
00:19:08
Speaker
And because of the separation, that's a lag, you know, we're creating this artificial lag of having data in one database and having to move across to another database.
00:19:18
Speaker
It's not needed.
00:19:19
Speaker
You absolutely don't need it.
00:19:20
Speaker
But the regulators,
00:19:22
Speaker
know i've i've seen even back at scdm you know regulators up on stage were saying no that's the way it's got to be it's got to be under control um so i think like you know it's not i'm not going to point the finger at the regulators i think it's also the vendors need to push the boundary and have a mechanism to communicate and voice their challenges to that and
00:19:46
Speaker
you know, that and for the regulators, they need to be listening to some of the new vendors that maybe haven't got a vested interest in keeping things the way the same way they are.
00:19:56
Speaker
Yeah.
00:19:58
Speaker
And be able to tell what's new and thought worthy and worth pursuing as opposed to maybe what's old and just that's the way we've always done it or that's the way the systems currently work.
00:20:09
Speaker
They need to be able to see what you can do going forward.
00:20:13
Speaker
Yeah.
00:20:13
Speaker
Yeah.
00:20:14
Speaker
I agree.
00:20:17
Speaker
So going back to, cause we've, we've went on a tangent a bit, which was very good, but going back to like the life cycle of clinical data, what would you like to see evolve in our industry?
Challenges in Data Cleaning Processes
00:20:32
Speaker
So, so yeah, so I think we, we have a problem today in the way in which data is
00:20:41
Speaker
pulled together and then cleaned.
00:20:44
Speaker
And the best way to describe it is one specific example.
00:20:48
Speaker
So let's imagine data is captured into EDC, a traditional EDC.
00:20:53
Speaker
I won't point fingers at any one particular system, but it's captured into EDC.
00:20:58
Speaker
And that EDC, let's presume it's got edit checking.
00:21:01
Speaker
So check the quality of the data using a rule and edit check.
00:21:04
Speaker
And then maybe it will have some mechanism for doing source data verification, then you can lock it.
00:21:10
Speaker
And many of the systems work along the same basis that you can only lock it once you maybe completed, you know, close your queries or done your source data verification.
00:21:20
Speaker
Now, some of the problems we see is that the fact that the data representation tools within EDC are not sufficient and therefore you take the data outside, you put it into some other tool.
00:21:33
Speaker
And what that then has the effect of doing is saying, actually, I can't press lock.
00:21:39
Speaker
because I don't know if I can press that lock button on that data point or that page or that patient or that site because data is being reviewed externally.
00:21:49
Speaker
Yeah.
00:21:49
Speaker
So it's been removed, right?
00:21:52
Speaker
That's just a tiny example, but it's pretty fundamental.
00:21:57
Speaker
If you're using a system that's designed to help you understand, right, the study is ready.
00:22:03
Speaker
If data has been worked on or processed outside in a separate tool or separate database,
00:22:08
Speaker
the possibility to do that is gone, right?
00:22:11
Speaker
It's broken.
00:22:12
Speaker
So you're going to have to say, actually, I'm going to revert back.
00:22:15
Speaker
What I'm going to do is I'm going to capture this.
00:22:18
Speaker
I'm going to extract it into an Excel spreadsheet.
00:22:21
Speaker
I'll do it.
00:22:22
Speaker
And then I'll use formulas to work out whether it's already, right?
00:22:27
Speaker
That is what happens in most clinical trials today, in my view, right?
00:22:32
Speaker
I don't have evidence, but I suspect looking at how the tools work,
00:22:36
Speaker
That's what's most often being done.
00:22:38
Speaker
Right.
00:22:39
Speaker
So I don't like the fact that data has to be taken out, given to something else because the original tool is not fulfilling the function that's required.
00:22:49
Speaker
You know, and I don't fully believe that it cannot.
00:22:52
Speaker
I think it just isn't right.
00:22:54
Speaker
And it's not recognized that it isn't.
00:22:57
Speaker
And there's vendors out there saying, look, let me fix this.
00:23:01
Speaker
You know, I will fix that over here because you guys can't do it right.
00:23:07
Speaker
which is fine, you know, we're cobbling our way to a solution, but it's almost the EDC system's got about 80% of the way there.
00:23:15
Speaker
or 90% of the way there, but the last 10% it can't do, which almost breaks the first 80% because you can't actually use the full life cycle anymore.
00:23:26
Speaker
So that right now is a bugbear because it gets you back to what I said originally, this lag, it creates a lag and it's lag for all the data because you're going, if these data points are wrong, can I use that patient?
00:23:42
Speaker
yeah so for the whole patience bad you know the whole patient is not in a questionable state um and therefore you gotta wait for all this to be clean so i i think um i think there's there's better work more logical i think it's more real world work you know when you lift the cupboard and see an excel spreadsheet used for data view
00:24:06
Speaker
you know, the developers, the systems, you know, vendors should sit down and go, why is that?
00:24:11
Speaker
You know, what do we need to do to take away the need that that user genuinely has?
00:24:17
Speaker
Yeah.
00:24:18
Speaker
Spreadsheet to do the job.
00:24:20
Speaker
Yeah.
00:24:22
Speaker
Well, it's going back to talking to the users.
00:24:25
Speaker
Hmm.
00:24:26
Speaker
It's interesting how I think I can see a lot of reproduction of what's been done already without too much thinking.
00:24:35
Speaker
And something I'd like to see evolve at my end is instead of saying, okay, an EDC, for example, does this, and that's all the EDC does, going back to changing
00:24:56
Speaker
what we think of an EDC because since we created the EDC concept in the industry, things have evolved.
00:25:07
Speaker
And like you mentioned, sites having more control of their data.
00:25:12
Speaker
We see sites starting to select their own technology.
00:25:16
Speaker
Or maybe there's better technology that would serve sites better in terms of collecting data.
00:25:26
Speaker
And then the sponsor needs to see that data, but they don't need necessarily an EDC.
00:25:33
Speaker
I think at the moment we're stuck in, oh, the sponsor selects everything, so let's not change it.
00:25:39
Speaker
Yeah, I mean, I think I feel for the sites.
00:25:42
Speaker
I think the sites are really hard done by, you know, this is discussions I've had and Brad Heiter has raised.
00:25:49
Speaker
So I think it's been very useful to have that dialogue and we have to keep having that dialogue.
00:25:55
Speaker
You know, this the principle of, you know, an investigator site that's got 10 clinical trials are running.
00:26:02
Speaker
and they have the potential of 50 different systems, you know, it might not be 50, but let's say it's 40 different systems.
00:26:09
Speaker
If you're a clinical research coordinator sitting down in front of your computer at nine o'clock in the morning,
00:26:14
Speaker
what the hell do you do?
00:26:16
Speaker
Which of these 40 systems do you go to?
00:26:21
Speaker
And many people would say, oh, if we just solve the problem with single sign-on, everything's gonna be great.
00:26:27
Speaker
No, you're still going to have to go to 40 systems.
00:26:30
Speaker
Yeah, you might skip the username and password problem,
00:26:33
Speaker
but you've still got to decide which of these 40 systems have got the most important query or the important signature to be applied.
00:26:41
Speaker
So you're going to start at 9 o'clock, you're going to work until 11 until you find the 37th system that's gone, God, yeah, God, yeah.
00:26:50
Speaker
And to me, that's just crazy.
00:26:53
Speaker
We've got to solve that problem.
00:26:54
Speaker
So what these sites are doing, and you can't blame the sites in any way, you know, Dr. Fox, he's a strong advocate of...
00:27:02
Speaker
on LinkedIn of sites owning their own technology.
00:27:06
Speaker
But in my mind, that's just growing to the problem.
00:27:10
Speaker
Yes, it might make things easier for the site, but it's making the overall map of technology and data that bit more complex.
00:27:20
Speaker
So the whole solution, if you take one clinical trial with all the sites, all the stakeholders,
00:27:25
Speaker
If you add on additional site technology in that sort of form where instead of having one EDC system effectively, you could add the number of EDC systems that you have as number of sites to the whole mix, it could be a complete car crash.
00:27:42
Speaker
But again, I don't blame the sites.
00:27:44
Speaker
They're going, this is unworkable.
00:27:46
Speaker
We need a solution.
00:27:48
Speaker
If the sponsors aren't going to fix it for us, the vendors aren't, we're going to take things into our own hand and we're going to fix them ourselves.
00:27:54
Speaker
Yeah.
00:27:56
Speaker
Well, I'm sure we could talk about this a lot longer, but that's all we've got time for today.
00:28:04
Speaker
Unfortunately, I've got one last question for you.
00:28:08
Speaker
What's the best piece of advice you've received that you apply to work?
00:28:17
Speaker
I've been thinking about this.
00:28:22
Speaker
I've got two actually.
00:28:25
Speaker
Never argue with an idiot.
00:28:30
Speaker
And actually, I should actually let me just get the exact one because it was marked.
00:28:34
Speaker
We never argue with an idiot.
00:28:35
Speaker
They'll drag you down to their level and beat you with experience.
00:28:41
Speaker
But I won't pick on that one.
00:28:43
Speaker
My favorite one was with the first software developer I worked with.
00:28:49
Speaker
It was more the advice was you should spend 50% of your time designing your software and 50% developing it.
00:28:58
Speaker
Now, design today is maybe different.
00:29:02
Speaker
In the old days, design would be different.
00:29:05
Speaker
But now, to go back to your example, you can actually use products like Figma to effectively interactively design a user experience with the user in front of you.
00:29:17
Speaker
So you're not actually writing the software, you're actually writing the design interactively with the end user and get them to grasp it and go, yes or no, that won't work.
00:29:30
Speaker
And you change it and you change it.
00:29:31
Speaker
To me, that's a design process.
00:29:34
Speaker
And to me, that should be 50% of your time.
00:29:37
Speaker
So the product management, the software design process is changing significantly because you get these super duper tools that will actually create what is a mock product.
00:29:47
Speaker
to make it look like it's a real product.
00:29:49
Speaker
And you take the user all the way through the life cycle.
00:29:51
Speaker
So you're bridging that gap that we traditionally had is you go all the way through software design, write it, deploy it.
00:29:57
Speaker
You know, a year later, they'd come back and say, oh, that's not going to work.
00:30:01
Speaker
And you go, oh, oh, back to drawing board.
00:30:03
Speaker
But now we've got rapid iteration, you know, wireframe tools.
00:30:06
Speaker
And, you know, I use Figma, but there's other tools out there that allow you to do that.
00:30:10
Speaker
So that is, spend your 50% on the design
00:30:14
Speaker
And then you won't waste your software development time producing something and the lag producing something that actually not solving the problems.
00:30:22
Speaker
Yeah, and it goes back to what you were saying about working with users as well.
00:30:26
Speaker
Yeah.
00:30:27
Speaker
And speed of development.
00:30:30
Speaker
Great piece of advice.
00:30:31
Speaker
Thank you, Doug.
00:30:33
Speaker
It's been nice talking to you as always.
00:30:36
Speaker
And I'll interact with you on LinkedIn, I'm sure about that.
00:30:42
Speaker
And I'll see you soon as well.
00:30:44
Speaker
And thank you everyone who's been listening to us today.
00:30:48
Speaker
You can find more clinical data talks on our website.