This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

Today on This Week Health.

The United States health care sector accounts for almost 8% of all greenhouse gas emissions. There's a variety of ways that health systems can improve their sustainability. One is by reducing waste. One is by reducing carbon dioxide emissions. And so that's one way that we can utilize informatics to reduce greenhouse gas emissions in the US.

Welcome to This Week Health Community. This is TownHall a show hosted by leaders on the front lines with interviews of people making things happen in healthcare with technology. My name is Bill Russell, the creator of This Week Health, a set of channels designed to amplify great thinking to propel healthcare forward. We want to thank our show sponsors Olive, Rubrik, Trellix, Hillrom, Medigate and F5 in partnership with Sirius Healthcare for investing in our mission to develop the next generation of health leaders. Now 📍 onto our show.

Hello everybody. I'm Jake Lancaster. I'm an internal medicine physician and the chief medical information officer for Baptist Memorial healthcare based out of Memphis, Tennessee. And today I'm excited to talk with Priya Ramaswamy about the sustainability informatics. Priya,`` welcome to the program.

Thank you Jake, for having me.

Priya. Can you just tell the audience a little bit about yourself, your background and what you do?

Yes. So currently I am a second year clinical informatics fellow to university of California, San Francisco. Originally I'm from Maryland, but I studied electrical engineering and computer science at MIT. I got my bachelor's and master's from there, went on to medical school at Tufts university school of medicine. And in 2020, graduated from my anesthesiology residency at Beth Israel Deaconess medical center, which is a Harvard teaching program.

That's great. That's a fantastic background and very excited to have you on. So I'm fairly new to sustainability informatics and by new means, this is one of the first times from hearing about it. So can you tell me just what it is we're talking about?

Clinical informatics, is a broad field where we use it and technology to improve the healthcare system. And one way we can use informatics is, and helping health systems improve their sustainability efforts. Given that informaticians have expertise in the EHR in clinical decision support in data science, there's many angles where we can help. Health systems improve sustainability measures with the using technology and data science as a focus.

And when you say improve sustainability efforts, what sort of efforts are we talking about? What, sort of things are we measuring?

Right. So there's a variety of ways that health systems can improve their sustainability. One is by reducing waste. One is by reducing carbon dioxide emissions. And so those are some ways that we can help with sustainability. The United States health care sector accounts for almost 8% of all greenhouse gas emissions. And so that's one way that we can utilize informatics to reduce greenhouse gas emissions in the us.

Okay, so this is fascinating. I was also in a, in a meeting recently where it was announced that the United States is going to start requiring hospitals to report on their CO2 admissions, I believe along with other other industries. But so it sounds like this movement that you're a part of is going to help kind of drive some of that.

That's correct. I've been fortunate that during my fellowship, I was introduced to this area because prior to starting my fellowship I never thought about CO2 emissions and waste as much in the hospital setting, but I was introduced and taken under the mentorship of Dr. Sima Gandhi, who is our sustainability director at UCSF. And so with my background in technology and informatics, I was able to help our institution try to eliminate or reduce the CO2 emissions that come out of our operating rooms, for example.

So that's fascinating. Can you tell us a little bit about your project and what you were doing?

Yeah, so just for a little bit of background. So the operating room, our ORs, one of the most wasteful settings in the hospital especially due to the use of volatile anesthetic gases. So anesthesia providers use volatile anesthetic gases for their ease of use in anesthesia and putting the patient under anesthesia, their speed of onset and safe.

However, the commonly used volatile anesthetic gases, mainly CBO flooring isoflurane. And does flooring have the global warming potentials that are more than 300 to 3000 times greater than carbon dioxide in a 20 year time horizon? So there was a clinical decision support tool done at UCF in 2018.

That reminded anesthesiologists to reduce fresh gas flow rates, which is the carrier for these anesthetic gases during surgery. So it would remind clinicians to reduce their rates of fresh costs for us. During the maintenance phase of anesthesia. And so during my fellowship at UCLA, I was appointed the technical lead to then expanded this clinical decision support tool across all of the UCS, namely UC Davis, UC Irvine, UCLA and UC San Diego.

Okay. So you have these anesthetic gases that are used while the patient's in surgery and these gases are what's putting the patient to sleep. Is that correct?

Yeah, so, so it helps have the patient under general anesthesia during the case, we use other agents often to induce the patient, usually IB medications, but generally these gases are what for a majority of cases are what keeps the patient under the maintenance of anesthesia.

Okay. And so why would a anesthesiologist keep a patient at a higher flow rate for those gases? What would be the. Is there any harm, I guess, to reducing that flow rate.

So, yeah. So historically anesthesia providers did avoid using low, fresh gas flow rates, namely due to safety concerns. In the past we used older types of CO2 absorbents in our machines and some of these anesthetic gases would develop a by-product that did have some theoretical. And actual safety concerns for patients. However, over the years, and improvements with technology, the chemical structure over a CO2 absorbance has vastly changed.

And so we don't have these by-product concerns. They're no longer. Concern. And so thus there has been substantial research showing that low, fresh gas flow is a safe and effective in one anesthetizes the patient and to being safe for the patient. However, as when people practice.

A long time in a certain way, it's difficult to change behavior. And so changing clinician and institution level behavior can be difficult even with just education. And so that's where informatics can come in and smart clinical decision support to help guide and change the behavior of anesthesia providers.

Okay. So because of historical patterns, they kept the higher flow rate and then. Your team built out an alert that was able to help them reduce those flow rates. Was there anything else you did? You mentioned education did y'all educate the the anesthesiologists that this was coming beforehand.

Right. So prior to my arrival at UCF in 2018, this CDs was developed in our epic software and launched and validated also. And so at that point there were several education sessions and grand rounds. Discussing the aspects of low flow anesthesia and why this is sustainable for the health system and improves the environment too.

But now we at UCF right now we run the lowest fresh gas flow rate in the country, a 0.7 liters per minute. When you ask other institutions across the country, oftentimes they run their fresh gas flow rates, much higher. So for example, where I trained residency for CBO flooring, we always ran her fresh gas flow rates above two liters, two liters or above permanent.

Even though we have. Modern CO2 absorbers during the time that I was there in residency. And so we had this tool that was validated and effective at UCF. And so how can we go about, and. Expand that across all of the UCS. And so yes, education was a big factor, but also gaining stakeholder buy-in and gaining physician champions and informatics champions at each of these institutions.

And how we went about doing this was starting in the beginning of 2021. Developing a. Multicenter team sponsored by the university of California office of the president to one educate the champions at each of these institutions were already very passionate about sustainability efforts.

And then through a series of meetings, educate them on how we developed our clinical decision support tool, and then helped them. Figure out how they can also launch a similar clinical decision support tool at each of their institutions.

That's great. So it sounds like you were able to make a lot of, or make a big dent in the problem of these admissions. Can you share some of the data or some of your outcomes?

Right. So at UCSF, there was a study that's to come out by Andrea Olmos and see Mugen DHL who validated our fresh gas flow BPA or best practice advisory. That's the epic lingo for this clinical decision support tool. And they validated this study and showed significant reductions in the amount of Seba flooring, for example, used within the operating rooms. I think they were able to show more than $150,000 in savings over their study period of a year due to the reduction of CBO flooring. And in addition substantial reduction in CO2 emissions. at UCF, for example, we used a non interrupted, active clinical decision support alert.

So non interrupted, meaning there would be a signal that would alert the clinician in the anesthesia information management system. For us, it's epic on the sidebar. But not disrupt other workflows in epic. If for example, there was a patient emergency and it was active CDs because it was running every minute during the case.

So one of the challenges. Especially when we were working with other UCS was how do you integrate hardware and bring in these signals into the software? Some of these data points, for example, were not always being collected before. And so. We were thankfully in successfully able to educate and train the informaticians and engineers at the other institutions in order to make it so that they can build a similar clinical decision support tools in their operating rooms.

Okay. So you mentioned the cost reduction, which is fantastic. Were there any countermeasures that y'all looked at to make sure that the alert or the reduction and flow wasn't causing harm?

Yeah, so, number one safety of the patient is our number one priority. And so some measures we took were that we didn't allow the alert to fire until procedure starts happened. And so our institution procedures start is like the actual surgical start of the patient where the surgeons do the timeout and the first incision is. For many cases you have like 10 to, up to sometimes an hour of time where the patient that gets prepped there's lines and things that happened to the patient Prior to procedures start and where a lot of hemodynamic fluctuation can happen, especially after induction. And so our alert does not fire until procedures start to give time for the clinicians to settle the patient get the patient to the appropriate anesthetic level and also not disrupt the other activity that may be happening in the operating room to prepare the patient for a second.

So that's one and then two hour alert can be snoozed. So in case there's something happening in the operating room, which does require a higher flow rates or hemodynamic instability you can turn off the, either turn off the BPA or snooze the BPA. And so we really wanted to make sure that We weren't interrupting patient care as much as possible.

Okay. And so you mentioned taking it from UCSF and trying to move it to. Other institutions within the UCs, what were the biggest challenges with moving it to another hospital?

So one of the biggest challenges was buy-in from each of the institutions. I think all of the institutions well, you see, why did there's a goal of sustainability and reaching carbon neutrality.

And so. The mission everyone agreed with, but then the challenge was sharing our BPA or alert system and having buy-in from clinicians because some of the biggest worries as I had mentioned, Was patient safety and disrupting peri-operative workflows. But with the series of lectures that Dr.

Gandhi and each leader from each institution gave to their respective institutions, we were able to get the buy-in. The second was just Standardizing and the data reporting, the hardware and software integration and creation. That was a bit challenging. And we worked over months and one of the challenges also was priorities of each informatics teams.

So in 2021, for example, there were still peaks of COVID cases. And so. There were priority shifts in the informatics divisions at each hospital, even in our hospital. And so carving out the time for the informaticians to build something like this in their EHR was challenging, especially when there was a pandemic happening, but we had such enthusiastic and excellent team at each institution.

So we're really proud of that. All of the institutions were able to launch the BPA.

Great. What about any technical challenges? So was it the case that every one of your hospitals where the, I guess the anesthesia machines talks with your EHR or did that interface have to be recreated?

So yeah, so technically there are some anesthesia machines that Maybe some of the older ones are certain brands that do not. For example, output the volume of gas used. So for example, like many of our machines that use CSF I'll put the milliliters of CBO flooring utilized. There are some machines that don't do that. So it did require that each of the UCS had to inventory what types of machines that they used.

Thankfully, most of the UCS have pretty modern machines. So there's only a handful of locations that use machines that this BPA would not be compatible with. But the vast majority, thankfully the machines were compatible with. There are some locations where, for example, even at UCSFs, the BPA doesn't can function, but maybe doesn't collect all of the information that we need for post BPA data reporting and analysis of behavior change.

And that may be in locations like MRI or some of the radiology suites where we do have to use Is special type of anesthesia machine, for example, because of like MRI compatibility. And those may not always output the type of information that we want to capture in our EHR.

That's great. What about any challenges related to the data side of things is all of this information already in clarity and. Yes then. Yeah. Is it I would imagine you all, aren't on one instance of epic. So you would have a lot of different clarity databases where if you're going to try to see the benefit of the entire project throughout the UC system how would you go about that? Would you have to aggregate it from all the different systems?

Yeah. so that is a challenge that we're working through right now. Thankfully all of these institutions do use epic even though they all use different versions of epic. And so for the listeners out there clarity is Epic's database or data warehouse that stores.

Almost the majority of information collected in ethics, EHR. And so you're right. So some of this information was not one collected in the EHR. For example, I think most EHR is we're collecting fresh gas flow rates and we're collecting the CBO flooring dial settings. Because that's part of the anesthetic record.

It's like how we anesthetize the patient. How are some of the information. Our databases. We're not collecting where the volume of CBO flooring used, which doesn't necessarily have to do anything with patient care. Because it's really what the volume that the machine used to spit out. There's different ways of calculating that if you don't have that value.

But it's not something that we were actively storing, for example, in our clarity databases. So the. Analysts and software engineers at each of the institutions did have to start pushing that information into their clarity database, which was not done prior. So that, that type of information we tried to get in before the BPA or the alert even launched just to have some data before the CBS was launched.

And then secondly even though all of this information is launched, there were some some of these institutions took Different approaches to creating their clinical decision support tools. So for example, you CSF developed a non interrupted, active BPA. Whereas one institution developed a very passive BPA, which just was a flow sheet color change on like when the fresh gas flow in increased to above one liter per minute, it would change from block to ride for.

The co the numbers, whereas another institution developed a interrupted, active BPA. So you would actually have a pop-up window of that. You had to exit out of before you can interact with other elements in the EHR. And so we're still figuring out how to best now, now that the alert is active in all of institutions, how to best Analyze that data.

Thankfully, we are in the process of submitting IRB fees through all of these institutions. And we feel pretty confident that we can accurately analyze the data post CDs for many of these institutions, because we've already been able to successfully demonstrate that at UCSF.

That's really interesting.

Different places to just have a different approach to BPA and. Analyze them to see which method was the most effective, almost like AB testing. We don't get to do a lot of that with them.

So and what's interesting is that this wasn't the plan. When we first started in 2021, we really thought, okay, like we develop like. Tool exists at UCS app, that doctor had gone the David Rabinovitz, the peri-operative director of informatics at UCF developed. And that Andrea OMA was validated in a post-study. so we really thought that the other UCS would take the same tool and copy it exactly. But every institution has their own needs and likes to run things their own way. And so by coincidence, different institutions decided to do slightly. Different approaches in how the alert would be shown to the clinician. Most of the institutions still use the same rules in firing however, how it appears to the clinician is different. So this is from informatics standpoint very interesting. We're very excited to study how these variations and CDs alert to appearance. Can effect clinician behavior and modifying change up reducing fresh gas flow rate.

Oh, that's, that's fascinating. And I think your project's fascinating. It's certainly an area. Of waste that I had never thought of. Is there anything else that y'all are working on in the future as far as waste reduction? Sustainability goes?

well, one project that I'm completing from this past year, so every year, the anesthesia department along with many departments at UCSFs have. Department wide quality improvement project that is resident led and being the in clinical informaticist fellow for the department I was in charge of helping guide the residents in their project.

And so given this passion about improving the sustainability of our healthcare system, the team decided to choose. To reduce like waste in the operating rooms. And so for example, like the operating rooms are some of the most polluting places in the, or because we have a lot of waste that occurs with each surgery.

Some often there's unused equipment, that's open. There's a lot of plastic products that are used at. Majority of them are single use if not all of them. And so there's a way there is a drive to reduce waste from the operating room. But another angle of our project was also how to reduce costs of certain high cost medications that we use in the anesthesia realm, where there are lower cost alternative.

And so I'm using some of my data science skills and teaming up with the database and teams that do inventory management at UCF. I was able to help the residents track per month, how certain high costs single use plastic products, how we can reduce the use of those things. When there's cheaper alternatives. Lower carbon footprint alternatives too. So one area we focused on was hotline use. And so historically UCSFs we do a lot of major cases where hotlines are definitely appropriate. Cause we were transfusing a lot. But sometimes hotlines were set up in cases where the EBL was like 10 CCS.

So there are some times where wasted hotlines that at our institution are pretty yeah. And I'm being spiked and set up, but never used or touched during the case. And so that naturally has to go in the trash because it's been contaminated environment. And so we had a major drive to reduce hotline use in the past year and were successful.

And that really measuring those, I think that's where a clinician clinical informatician comes in with the data science and learning about like inventory manager. And tracking those things. I think we definitely have the skills to help with those types of initiatives. A second initiative that we did was also our department realized that plasma light was more expensive than lactate and rigorous at our institution.

And a lot of cases were unnecessarily using. Plasma light for her outpatient surgeries, for example, when the need wasn't necessarily there. There's definitely times when plasma light is appropriate to use for the patient. But for many cases, it was inappropriately being used and costing the health system and the patient also.

And so we were able to track both with clarity data. Where we document all our fluids given in drop and also warehouse inventory management on the per day level, how much plasma light versus LR was used. And so by the reduction of plasma light and increase in LR, we were still able to save money because plasma light was significantly more expensive when you look at the costs between LR and P light.

So that's one area and then also. Reducing some of the expensive medications where there were clearly cheaper alternatives at our institution. And so I think that's an easy and great way for other healthcare systems to team up with informaticians who have both the clinical insights know what's safe for the patient, but also the data science and clinical informatics background to team together and try to use data-driven and tech driven solutions to improving healthcare sustainable.

No. Yeah. I was going to say the same thing. Just that unique blend of being an anesthesiologist, understanding data science and informatics, you've just rattled off several things there's likely to save your institution, hundreds of thousands or millions of dollars. So that is as well as be great for the environment. So that is a very great skill set to have. And so glad we were able to highlight it here and thank you everybody for.

Thank you so much, Jake for having me.

I love this show. I love hearing from people on the front lines. I love hearing from these leaders and we want to thank our hosts who continue to support the community by developing this great content. We also want to thank our show sponsors Olive, Rubrik, Trellix, Hillrom, Medigate and F5 in partnership with Sirius Healthcare for investing in our mission to develop the next generation of health leaders. If you want to support the show, let someone know about our shows. They all start with This Week Health and you can find them wherever you listen to podcasts. Keynote, TownHall, 📍 Newsroom and Academy. Check them out today. And thanks for listening. That's all for now.