- Well, welcome to the Friday lecture series in Architecture for Health.

It's great to have you with us and you're in for a special treat today.

Our guest today is Dr. Roxana Jafari.

Now, she just recently joined the Department of Architecture here at Texas A&M and we're excited to have her.

She's an assistant professor in architecture and her specialty is healthcare design research.

Her primary focus centers on enhancing patient safety and overall outcomes within high stress healthcare environments.

Prior to joining us here at A&M, she was an assistant professor at Lawrence Technical University in Michigan, and she also served as a research assistant at Clemson University's Center for Health Facilities Design and Testing.

And I think she's going to mention some of that today during her remarks.

Now, as you know, this series is about frontiers in research that might impact evidence-based design.

Let me just give you a peek at some of her research interests, then maybe that sets the stage for what we're going to hear from her today.

Areas of research interest for Roxana, high risk health care environments, technological based interventions in health, tool standardization and development, environmental quality for aging populations, and geocoded or geographic information systems to understand health related issues.

That's a pretty broad palette.

So I'm excited to welcome Roxana to the Friday lecture series in Architecture for Health.

Please help me welcome Dr. Roxana Jafari.

(audience claps) Welcome.

- Well, thank you so much, Ray.

Well, howdy.

- [Audience] Howdy.

- Thank you so much for having me for the lecture series.

It's a pleasure to be here.

Today I want to discuss evaluating design during the design process.

And I'm gonna talk briefly about the human-centered design approach that you might have heard about.

So a little bit about my background and of course, Ray covered it really well.

I got my PhD at Clemson University.

I got it in architecture and health program and I worked briefly with HKS as a health research intern.

Worked with the research team.

And I also worked for two years as a tenure track assistant professor at Lawrence Technological University where I had research assistants and we were constantly working on research projects.

So my area of research, what I use in my work is normally related to evidence-based design and human-centered design approach.

Are you familiar with human-centered design?

Yes.

So do you know what steps are included in human-centered design approach?

Like if you want to do a human-centered design, what is the first thing that you guys do?

- [Audience Member] Get IRB approval?

- Very good one, yes, if you want to engage users because it is basically focused on your users.

So maybe an IRB approval would be needed.

Very nice.

What else?

Okay, I have the summary of the steps here for you.

So normally the first thing that you set, the first thing that you do is to understand the context of use, like what is your project, where is it located?

And you need to collect as much information as you can on this.

And then you need to identify who your users are and collect information on them.

If necessary, get an IRB approval and collect data, interview them, collect data from them using surveys, to really understand what your users need.

And then you will develop design solutions and constantly, while you are following these three steps to found your work on human-centered design, you wanna do evaluations, you wanna keep evaluating your design to make sure that it is responding to the user's needs.

And if it is in line with whatever information you gathered regarding the context and the population who are going to use your basically project, the building that you're designing.

So in this presentation we're going to be mainly focusing on the evaluation process, what methods we use as architects, as design researchers to evaluate design and how we do this and also what issues we normally face with.

So the first project that I brought for you here is a collaborative project that we used, virtual reality and also laboratory experiments to evaluate one aspect of the design.

So in this project we were evaluating the lighting quality in an office workstation and we were interested in evaluating these daylight control options.

We had blinds and we also had electrochromic glasses that you can see on the right.

So we collaborated with a View Inc, who is basically a team working on the implication of smart windows and our impact on human comfort and health.

And we actually wanted to see which one of these daylight control solutions would work best in the office that we were designing.

And so in this study, we were not evaluating all aspects of design, like for example, space arrangements, adjacencies, it was just about lighting quality and that was the only thing that we were evaluating.

So let's see how we did it actually.

To do that, first we installed the daylight control samples in an basically lab environment located in Mississippi, in the US, you see the different options with different degrees of openness and different degrees of transmission of daylight.

And we place some sensors for measuring the light levels, for measuring the temperature.

And we also took some photography to see what is the degree of heat as light passes through these windows.

So basically we did some site measurements.

In addition to this, we also used the Matterport tripod cameras and we created a realistic, 3D imagery of the interior.

Our goal was to actually have this environment tested by users in person and also using virtual reality.

So this is basically how we evaluated the design using virtual reality.

We had people participating in the lab, we collected data and because we were, you can see that we were all wearing masks.

It was 2021.

It was difficult to get users to participate in research projects.

So in order for us to get enough users and get enough feedback to refine the design process, we had to actually use that Matterport camera, create that virtual environment and recruit more users across the US.

So this is our campus at Lawrence Technological University in Michigan, where we recruited more people.

And this is what the users experienced when they wore the VR goggles to experience the interior of the office environment.

So it was not a kind of virtual environment that we created using Revit, Lumion, so it was the actual imagery of the interior of the office with the actual product being installed.

So this basically 360 environment was created in the morning between 08:00 to 11:00 AM and this was when we were interested in evaluating those daylight control options, as you can see.

So we used the VR and as you can see in this picture, we also had participants coming to the lab in Mississippi sitting at each workstation and filling in a survey.

And while there were in the virtual environment, we also read the survey for them and we collected their feedback.

The questions and the surveys were completely similar.

One of the reasons that we used both methodologies was that we normally know that VR has some limitations in evaluating the space, but we were not sure what limitations we would face in evaluating the lighting quality.

And that's why we combined the evaluation in the lab environment and virtual reality to investigate it and communicate this with architecture, with people in the architecture field who are interested in evaluations.

And this is what we found.

So you can see that the upper graphs are evaluations in the physical environment and the lower ones are evaluations in the VR office.

So you can see there were some variations in participant's feedback when, you know, you look at the physical versus virtual space.

But what was really interesting is that despite this variation in the responses, the overall satisfaction with the type of daylight control option was kind of consistent.

You can see that both groups prefer the electrochromic glass compared to other daylight control options.

And this is a summary of all the findings with site measurements and the surveys in the physical versus the VR office that we collected.

And you can see again that there was a good level of agreement.

However, there were some limitations that I want to bring to your attention.

One of the things that we found was with severity of glare.

So for example, for some daylight control options like this one, we found actually, we found out that this option would cause perceptible glare when we did onsite measurements and when we tested this with actual participants in the lab, while when we presented it to the people in the virtual environment, they noticed that there was some glare, but they mentioned that, well, the glare isn't too bad.

We think that we would still enjoy daylight.

So this was something that we found by this study.

And I want you to be careful when you are evaluating daylight using VR, just note that daylight evaluation using VR would minimize or underestimate the severity of glare.

So if you want to use VR for evaluating daylight quality, maybe you would need to combine it with other evaluation methods, like laboratory experiments and so on.

So that was just one simple study and evaluation method for looking at daylight quality and daylight control options.

So just one aspect of design.

I brought this project to you, which is more comprehensive in terms of using evaluation during the design process.

And this project is for, oh sorry, let's go back.

I think my pointer is not working.

- [Crew] It may not show up on the glass.

- Okay, here we go.

Alright.

So this project was conducted at Clemson University when I was a PhD student and a research assistant at the center.

This was regarding improving safety in an operating room.

So it was all about redesigning operating room environment.

The PI for this project was Dr. Angelique Joseph.

And we did a collaboratively with a large multidisciplinary team from Medical University of South Carolina in Charleston, and also researchers from different departments at Clemson University.

So the goal was to optimize the operating room space and then incorporate the optimized design in an outpatient surgery center that was supposed to be built in Charleston.

That was the ultimate goal.

So let's see how we basically did it, how evaluation took place during this human-centered design process.

And if I can move to the next slide, which seems like I can't.

Here we go.

I just need to, all right.

So the first thing that we did in order to understand how we can optimize operating rooms, we started by watching some videos of surgical procedures.

And as a research assistant, I was also part of the team and we monitored the surgical procedures and we coded the behavior of the surgical staff, second by second, to basically understand what they do, what their needs are, how they move throughout the space.

And you can see.

You can see for example, we monitored each surgical member, surgical team member separately.

We coded where they were located during the surgical process, how they moved throughout this space and this map, that this is slide that is constantly brought up, you can see that we even mapped their workflow walking path to understand how they moved throughout the space and if the shape of the operating room that they already had in their older facility was working well for them.

So it was all in their older facility that we were studying to see basically what is working for them and what is not working for them.

After doing this, the ideation process started.

So in collaboration with the master's students in the architecture and health program, we started ideations for the operating room environment and to refine the design strategies and and ideas, we use the physical mock-ups.

And I brought some videos for you to see how the physical mock-ups were actually created.

So, alongside these physical mock-ups that I'm going to show you very soon, we also use other methodologies to get feedback from clinicians involve them during the design process such as, for example, we shared traditional, architectural drawings with them, floor plans sections, we showed them some renderings, 3D perspectives, we had virtual reality.

But I want to focus on creating the physical mock-ups for evaluation in this presentation.

So let's see, basically how the mock-up was created.

So I'm gonna show you the previous video.

I'm having a little bit trouble with this, so bear with me.

Okay.

Seems like this is not working.

Thank you so much.

Alright, so this is the master's student team working on building the mock-up.

So they're putting the walls together.

They started with creating the boundary, taping the boundary of the operating room on the floor in the lab space that we had available in Charleston.

They set up the physical mock-up.

They did an amazing job in creating this in one day.

So it was a low fidelity cardboard mock-up that we created at the initial phase of design just to refine the design strategies and design ideas.

And after the walls were put together, the boundaries were created.

I'm gonna move to the next video.

The clinicians who were supposed to occupy the operating room that was supposed to be built also came into this operating room.

And as you can see, the team could move the pieces around, change the location of the doors and the team would actually simulate different phases of surgery, bring in the patient bed and practice different workflows to see which location of design elements, which arrangement of design elements would work best for them.

So this was just low fidelity mock-ups to understand what is working for the surgical team and what is not working.

After this phase, we switched to higher fidelity mock-ups.

So we had actual structures, surgical booms, we had the screens and we had high fidelity mannequins.

So this is a simulation for pediatric surgeries and the same surgical team that was supposed to occupy that surgical facility, the future operating rooms, they participated in the high fidelity mock-ups, they simulated the whole surgical process.

And then after that they shared their feedback with us.

So the goal was for us to see if di design elements and details, like for example, the material on the floor, the location of the screens, the booms, the location of the equipment at the anesthesiologist side, everything was working well for the surgical team before this operating room is actually built.

So after each of the simulations, we held debriefing sessions with the team to get their feedback to basically understand what would work best for them and what we needed to change.

So we were constantly evaluating design before we made the final design decisions for this operating room environment.

And eventually, if I can show you the final picture of the built operating room, this is the finalized finish operating room constructed in Charleston and of course currently it is occupied.

The team also conducted a post occupancy evaluation after this operating room was occupied for six months plus.

And this is still working on refining the design process.

So after the project is complete, the evaluation process does not stop.

We usually keep collecting data to ensure whatever was implemented and constructed can actually work really well for the design teams because later after the project is constructed, some issues might come up that we never noticed during the simulations, mock-up experiments and simulations or the design process.

So some lessons learned that I want to share with you, if I can go to the next slide.

We use different types of media.

We use floor plans, 2D drawings, we use virtual reality, we use physical mock-ups to involve users during the design process and collect data from them.

And I just want to mention to you what worked best for the clinicians when they were getting involved in the design process and what did not work for them.

For example, the floor plans and perspectives, what you as design students normally produce, they were the least effective to communicate design ideas and strategies to clinicians and get their feedback.

They could understand what was going on, but when they participated in immersive VR experiences and when they were present in the physical mock-up, when they could actually move things around and change their environment, that was when they could actually interact with the design strategies and understand the design and help us really refine it and make it better.

So physical mock-ups were their top choice, followed by VR mock-ups.

And as you know, and as we discussed earlier, VR has always limitations and challenges in terms of getting feedback from the audience, from users.

And there are limitations regarding visual realism, communicating textures with, for example, lighting quality as we discussed previously.

And apart from that, some people are just not comfortable being in an immersive virtual environment for a long time.

Some are prone to motion sickness cannot tolerate it.

So there are always challenges.

But with a physical mock-up, you do not have a lot of challenges like this.

On the other hand, physical mock-ups are expensive.

Building them is very time consuming.

So you always need to, as designers and design researchers, try to think what is the best strategy for you to evaluate your design or what combination of strategies and methods you can use in a given time to evaluate your design, involve your future users and get their feedback to refine your design.

And talking a little bit more about post occupancy evaluations.

I want to bring to you another project that I also worked on in collaboration with a healthcare facility in South Carolina.

So this was regarding evaluating the ICU room layout in an ICU unit located in Anderson, South Carolina.

And as you can see here, this ICU unit had three different types of rooms.

The rooms in green were windowless, you can see the layout over there.

The rooms in yellow, they were windowed, but in a way that the patient bed was parallel to the window, you can see it as room type A.

The patient bed was parallel to the window and allowed the patients having access to outside views and daylight, both.

The rooms in pink, however, they had the patient bed perpendicular to the window with the patient head placed against the window letting the patient get daylight, but the patient could not see outside views because they were actually facing inside.

So we were basically trying to evaluate these three rooms after this unit was occupied for years to see how these different room layouts were impacting the patients.

And in this case we were interested in health outcomes and we were trying to relate the room layout to the patient outcomes.

Like for example, how soon they recovered and what the mortality rates were, what the delirium rates were when they were hospitalized.

So these are another view for you to see how the different rooms look like and how they were different in terms of incorporation of daylight and having access to outside views.

So these were some of the outcomes that we were interested in, which came up during our discussions with the team of clinicians, who wanted to conduct this study.

So as a researcher, the first thing that I wanted to do, I tried to get electronic health records data from the hospital to basically see what the patient's outcome were and how we could actually relate it to the design strategies that we were investigating.

So I thought to myself that, well I'm a researcher, design researcher, I understand design, I also know analysis and statistics.

So I would get that data from the hospital and they were willing to share it with me.

There were no complications in that regard.

We went through the IRB process and we did everything.

And I was really happy that I could complete this project as soon as possible.

But there are always unforeseen complications that happen in the research process and I'm gonna share that with you right now.

So when I got the data from the hospital, and by the way their electronic health records system was Epic Systems.

So if you are interested in research, it is good for you to understand if you want to actually work with hospitals as healthcare designers, what their healthcare system is, their method of recording data is what system they use.

For example, in this case it was Epic Systems and this is the output that they shared with me.

And this was when my challenge started.

So I looked at the data set and I noticed that everything was abbreviated.

So there were terminology that I was not familiar with.

I was interested in some outcomes, like for example, I was interested in mortality but there were absolutely no columns in the data set showing me what the mortality rates were or what the delirium rates were.

And I also knew that if we wanted to correlate some design strategies with outcomes, we also need to understand other factors about patients, like how sick they were.

Did they have any complications, any other complications like obesity, diabetes?

So you need to know this to ensure that whatever is causing the impact is directly related to the design of the environment, not other factors like physiological factors.

So that was really challenging because I noticed that and these variables do not exist in the system and it is up to me as a design researcher to go through all of these doctor's notes over here and find those variables regarding patients and code them and get them basically.

So there was a lot of work that needed to be done and we do not have the health records that would directly share with you whatever you need to know as design researchers.

So I started working with the data and I had multiple sessions with the clinicians who kindly helped me understand this data set and eventually we could actually conduct the analysis and the results were shared with the hospital.

Another thing that I want to bring up in terms of limitations, and I think it's really important, during the discussions I found one important variable which was called the severity of illness.

So it is a combination of multiple factors like the patient's age, patient's primary, secondary diagnosis.

It is a complicated algorithm that is very useful for us to compare outcomes across different designs for the rooms.

And I was really happy that I found this variable because then I could know that if I see differences in terms of for example, how long patients stay in a room, it was not due to their different levels of sickness.

It was basically the design and availability of daylight and views that was making the difference.

But what I found out was that well we had a considerable number of missing data that were never recorded in the data set.

And when I actually filtered that data, I realized that I was losing about 900 patients which limited my sample.

So we have these limitations when working with data and working with data is an important part of both occupancy evaluations if you're interested in it.

But eventually we could actually do the analysis and we could measure, for example, which room benefited patients most in this case room type A, which provided direct access to outside views was really benefiting patients.

And another thing that we were interested in was to understand symptoms of anxiety, depression, and even sleep quality and how that would be different across these rooms.

But again, the challenge that we faced was that hospitals based on hospital protocol do not automatically record that data.

So if as design researchers you are interested in that data, you need to go to the hospital monitor patients and record that data yourself.

So I did that for two years.

I went to the hospital and I monitored patients for two years.

I recorded their anxiety depression level on a daily basis and also sleep quality.

I used Fitbits to record their sleep patterns.

And we also noticed that well as expected rooms that provided availability of daylight, positively impacted sleep quality, anxiety and depression.

So that was some of the challenges that I had with that data set, which we eventually included about 2,500 data points in the analysis and we had about 3000 patients to look at in that timeframe.

The problem got a little bit bigger when we started working on bigger data sets.

Like for example, for this project, which was not at the facility level this time it was at a community scale.

And we were looking at the impact of travel time to the emergency department on heart disease outcomes, like how soon patients recover.

And we were looking at the data from 2011 to 2019 and you can see the geocoded location for patients traveling to that hospital.

And this time I was working with 25,000 data points and all the issues that I mentioned regarding going through doctor's notes and coding certain variables.

Now this time it had to be done for 25,000 patients.

So you can imagine that this process is really time consuming and labor intensive.

And as we are moving forward, we really need to think about some methods, some ways to expedite this process for design researchers and designers.

And of course finally we did it.

So it just took a lot of time, but eventually we were successful in cleaning, sorting the data and use GIS to code this and deliver this project to the hospital.

And it just takes us to these concerns related to the field research in this field, evaluation in design.

And these are the things that I listed.

And I think especially when it comes to post occupancy evaluations, these are the challenges that we are facing today.

So there are some concerns regarding liability and funding concerns, liability in terms of, well, you know, if you do post occupancy evaluations in hospitals and it turns out that there's a design factor that is not working well and negatively is impacting patient safety, there might be some issues for the hospital or the architects participating in designing that they might find it hard to secure future projects.

So there are concerns regarding that when it comes to post occupancy evaluation of healthcare facilities.

And sometimes it's hard for architects and clinicians to agree upon the factors that should be evaluated in the post occupancy evaluation.

And as I mentioned, if you as design researchers, you want to use data, electronic health records to see how design is linked to patient outcomes, there are complications, there's an issue of finding the variables that you want.

We have the issue of missing data.

So we have to find ways to overcome that problem.

And there are some issues regarding missing data, some variables that you might need might not be there.

And also the process is time consuming, and labor intensive, As I mentioned.

And this is the part that I want to actually hear from you, maybe some of you have been involved in post occupancy evaluations, maybe have heard about strategies, how we can make it better.

So these were my thoughts, I included them here.

I thought maybe to address this issue and make things better in the future for healthcare designers, architects for clinicians, maybe we need to do more of cross training courses.

Workshops that involve architecture and clinicians, courses like this.

The program that you're currently in is doing a wonderful job in training you and introducing these existing issues with you.

And if we do the same for clinicians, that would also be really helpful.

And I've also heard that some other programs that Texas A&M has, like the ESET program, which involves clinicians and engineering students, which is really paving the way for the future of this field.

And also evaluation of the design.

And also it would be great if in the future we could have data sets that could directly link design features and health outcomes and be automated to address, you know, all the necessary variables in the process.

And also it would be really helpful if we could integrate smart technologies in the healthcare facility environment that could automatically record like lighting, temperature, noise levels and the variables that we constantly work with to understand how they impact patients' health and safety.

And also maybe we could come up with some ways to overcome liability concerns and especially financial barriers.

Maybe if there were financial incentives for hospitals to run post occupancy evaluations and check their designs to ensure that it is positively impacting staff and patients, it would be really helpful.

And also we're living in an age that artificial intelligence is everywhere and we have been using that in the field.

We use different methods like machine learning, natural language processing, and still when you want to actually use artificial intelligence, it is really important that you deliver, you basically have a clean input for the machine, for the system to work with.

So how can we actually streamline this and make sure that again, the responsibility of the researcher is not that extensive.

So the AI system is automated and can actually run the analysis for us.

So these were my thoughts and I would like to hear it from you, like what issues, what other issues you have encountered when doing research, when collecting data during the design process, especially in post occupancy evaluations.

And how do you think we should address these issues to make the future better for designers?

Yes.

- [Audience Member 2] My question in regards to the virtual reality, you were mentioning lighting being somewhat hard to trust because of the unrealistic.

Have you thought of getting together with Unreal or Epic games?.

'Cause I know you Epic systems, but the games I know trying to get into mental care with precision.

- Yes, so I just want to mention to you what we did for this project.

So we could actually simulate light like in a virtual environment, like the way that Unreal systems do in video games that you might have played with.

But in this case, we wanted to actually record the imagery of the actual daylight passing through the material.

And we could actually simulate that in software like in Unreal Engine.

But we thought that the Matterport camera with actual sample and the actual material of the daylight control system installed in a lab would give us more accuracy, 'cause that way we had to actually program the exact material, you know, reflectance factor and everything related to those daylight control options into the system.

And that could take a lot of time.

And we thought that considering the limited time that we had and also availability of this technology Matterport, it could give us a better, more realistic input with the natural light crossing that, crossing through that material.

So even though we made an attempt to record the realistic images, we still faced limitations.

So it means that in the future we still need to work on this.

But we also did that.

And in my opinion, Matterport with actual materials and daylight could actually be more realistic.

Sure, yes.

- [Audience Member 3] Actually it's not only about evaluation, I actually have a partner about the gap between the research and real world practice.

So for me as a user, sometimes I receive the service about the service or the service or the denial space and I feel out, but I think they will not change it.

So it's like waste of my time.

But as a research researcher, while I'm doing research, I try to recruit many person people into my research and I try to let them know it's will benefit you in the end.

But sometimes there is always gap between the research and practice.

So I think there is like a gap between the users and researchers, how can we break that?

- So do you mean like especially regarding what you mentioned when you fill in the survey and you feel like they do not change the environment, in healthcare facilities once they're built and occupied, some changes are very difficult to happen in the environment.

It's very challenging.

It's gonna be very costly.

So that's why it's really challenging if you fill in the survey and you communicate with a provider that something like that patient room that you were in the environment was not working well for you.

So changing that would be very difficult.

They might be able to change some design elements like daylight control options, blinds, the interior furniture finishings.

But other than that, that's the challenge with post occupancy evaluations as well.

Some of the changes can be really expensive.

- [Audience Member 3] Yeah, I totally understand.

It's like we try to do our best to do the research and find the best result or better outcome for better service, better patient experience or better like working space.

But sometimes it's hard to be really practiced in the real world..

So this gap, like it's confusing.

- This gap exists, yes.

- [Audience Member 3] And sometimes as a user you will be confused like, I'm wasting my time.

They will not change it at all.

And sometimes in research I'm also confused like if we cannot apply this into the practice, what we're doing, like just do paper publication and it will work.

But it's never be applied in the real world.

- So what is really important for you as a designer?

First, make sure that you are communicating your design ideas and clinicians in a language that the clinicians understand because there is a gap in understanding, for example, the field.

So clinicians might not understand the design field, the designers might not understand the basically clinician's issues or certain terminologies.

Like the problems that I was facing with managing and sorting the data set.

I basically did not know certain terminologies and I did not know how the Epic system was set up.

So it took some time for me to work with clinicians and learn it.

So this gap leads to not understanding each other really well.

And we need to overcome this issue so that clinicians understand design, designers understand what clinicians need and they understand the work process is better to overcome this issue.

But there's a long way to go.

I agree with you.

Definitely.

- [Crew] The question in the back?

- Oh, yes.

- [Audience Member 4] Well this is kind of a thought on how to address current issues.

I'm thinking maybe forming a committee amongst architect clinicians and patients to kind of better streamline how to address those big issues.

So maybe that could be something that's implemented.

- Could, you elaborate more?

Like what should we do?

Like holding workshops for example?

- [Audience Member 4] Probably committee that's like amongst hospitals for example.

Like I'm a part of a community with patients, clinicians at Parkland.

And maybe every maybe four months to get together and figure out like best practice and stuff.

- Absolutely.

I agree with you.

These attempts are really, really good.

And whatever we can do to bring clinicians, to bring designers, bring the staff, everybody together to sit and understand and discuss what their needs are.

And the more they keep talking about these issues and try to actually come up with solutions, the better they can understand and the better they can address these issues.

We just need to hold more workshops, create more instances for people from different fields to come together and start working on these issues together to overcome that gap regarding communication.

Because that already exists.

- Roxana, what's the path forward if the standard data sets that the clinical fields use are built maybe originally for financial purposes, and then morphed into clinical purposes and they have a divided kind of history, of schizophrenic origins.

And we come as designers to the data set and say, we would like to answer this question, but in order to answer it, we need raw data that didn't belong in the financial and didn't belong in the clinical.

And where do we find a path forward that's not the extreme of rebuild the entire data set every time you want to ask a new question.

And having our hands tied because the raw data that we need is not routinely collected.

How do we find the path forward as researchers to get the data we need without asking the industry to completely retool?

- And we're not asking that.

So we just want to make, we ask them to make a small change in the existing tool and improve it.

So we're not asking the industry to create like a new Epic system from scratch.

So, and this is by the way, the issue that it's not just specific to the United States.

I was talking with another team from Australia this summer that they're facing the same issue over there.

That the design researchers, they're having a hard time getting data from hospital.

What they want is not being recorded based on hospital protocols.

And they are actually doing something to overcome this.

They're having, they build this multidisciplinary team, they are thinking what can we do with the existing system and how can we build on it so that it can actually help the field better in the future and include the variables that we can work with so that we as researchers and designers do not spend months and even years to just get the data that we want to link the design strategies with design outcomes, with health outcomes.

- Great.

Other questions, as our time is near at hand, any last question that someone would care to ask?

Well, our time is up.

Roxana, this was terrific.

Wonderful to hear from you.

It's wonderful to have you on the team.

Welcome again to our faculty.

It's great that you're now one of us.

We're delighted and excited about the frontier of your research efforts with us.

Thanks so much for being here today.

Help me thank her.

- [Roxana] Thank you.

(audience claps) (upbeat music)