Episode 26 - Dr. Porter - OB/GYN Risk and riskLD

How can OB/Gyn physicians use technology to reduce adverse patient events? Dr. Steve Porter of riskLD explains how to use software to make maternal care data actionable.

  • 00:10:11 - 05:10:11

    Welcome, everybody to reimagining healthcare a new dialogue with risk and patient safety leaders presented by Medplace. We're excited to bring you conversations with top risk and patient safety thought leaders from organizations across the country, please subscribe to get the latest news and content. And if you found this episode valuable, please share it with your colleagues to create some meaningful dialogues in your own communities. If you're interested in participating as a guest, please feel free to email us at speakers at medplace.com. And also check out our related content like articles, videos, and more on blog dot medplace.com/resources.

    My name is Jerrod Bailey. I'm the CEO and founder of Medplace I'm going to play host today. And today I'm joined by Dr. Steven Porter, CEO of riskLD, a health care or health tech startup, much like us with a focus on OB GYN risk. Welcome, Steve, how are you?


    Steven Porter 01:10

    Very well. Thank you for having me, Jerrod.


    Jerrod Bailey 01:12

    So I got to know you at a conference that we were at relatively recently. And I love I love learning about tech startups that are that are in this space that are that are solving problems around risk and patient safety. That's obviously where we're, we're embedded as well. And I was really impacted by what you had to share. You guys have some pretty impressive metrics. And you've built a tool, I think to help support what you're doing and kind of interested in unpacking some of that today.


    Steven Porter 01:43

    Absolutely. We're delighted to be sharing our story with you. Right?


    Jerrod Bailey 01:47

    Well, all right. So what is risk LD?


    Steven Porter 01:51

    Sure. So risk LD is a patient safety solution. It is a software platform that fully integrates with the inpatient electronic medical record. And it provides a combination of early warning and clinical decision support at the point of care on labor and delivery and postpartum. So the whole mission of the software, the mission of our company, is to improve perinatal outcomes, and then to reduce risk for laboring moms and their babies.


    Jerrod Bailey 02:18

    Fantastic. Well we sit sometimes if we're working with carriers or TPAs, we're sitting at the the other end of the far end if when things go wrong, and it becomes a claim, or it becomes a lawsuit, and this is just one of those specialties, that for lots of reasons that we all understand, end up generating a lot of cost for the entire system. Because when something goes wrong there's usually a lot of emotion behind a lot of other things. So anybody who's trying to affect this particular specialty is thinking, I think, is doing a great service for lots of providers out there. But give me a little bit of the backstory like a little bit of your backstory in the backstory to risk LD like how did you conceive of this idea where to come from, when you start to realize there was a need?


    Steven Porter 03:08

    Absolutely. So I am a practicing obstetrician gynecologist, I maintain an active clinical practice, see patients a couple of days a week, both in the hospital in the office. And the need for risk LD, the really was was spun out of about 10 years of quality and patient safety work at a large health system in Northeast Ohio, where I'm affiliated. And essentially, the team had been looking for many years at drivers of cases that have come up for quality assurance review, if ever there had been an adverse outcome for either a mom or a baby. And the team realized that almost any of these cases almost all of these cases could be traced back to one of two sources of error. The first being what we call a lack of situational awareness was essentially it just means that the clinical team wasn't aware of a problem until it was too late.


    The second problem that was seen more community hospitals as opposed to academic medical centers was patients not being managed optimally, patients not being managed appropriately even after a high risk condition had been flagged. And so we stepped back and we said okay, well at risk of oversimplifying the problem. If, if, if is this, this can be traced back to either not knowing about the problem or not managing it effectively. Wouldn't it be great if we had a technology that we could deploy on our labor and delivery units that would help us identify and hopefully mitigate these sources of error? And long story short, we couldn't find anything commercially available on the market that met our needs. So ended up designing our own for a grant through an Ohio State agency the lattice to develop an internally prototype and validate the software Er, and that technology was exclusively licensed from the hospital at the tail end of 2018. And that's what we're now commercializing as risk LD.


    Jerrod Bailey 05:09

    Amazing. Well, tell me like as you're as you're navigating the market, you're introducing this to different facilities, do you have an inkling for? What types of facilities maybe have the most to gain out of this? Or should be should be considering it first? Because some qualifiers there?


    Steven Porter 05:28

    It is a great question, I think it lends different value in different kinds of clinical settings. And I'll give you some examples.


    There are really busy, urban, inner city academic medical centers, that have high volumes of high acuity patients. And in that clinical environment, it's really the situational awareness piece, that's the biggest source of risk. It's not necessarily knowing what to do to manage the problem. Because many of these institutions have maternal fetal medicine, high risk obstetrics, in house or available for consult, it's more a question of, in that very busy constantly changing clinical environment, sometimes critical information falls through the cracks, there's a lab result that somehow gets missed. There is a fetal heart rate tracing issue that doesn't get attended to there's an abnormal vital sign that isn't reported, and don't know about a problem, you can't manage it effectively. So in these really busy, high volume centers, sometimes the biggest challenge is knowing where to look first, knowing where to pay attention.


    Now, you contrast that with some smaller community or critical access hospitals, where you might have a different set of problems, you might not have OBGYN providers, you might have family medicine providers, you might not have medical doctors in house, they might be remote or taking call from home in some low volume settings. Sometimes these types of hospitals don't see complexity, with sufficient regularity to be able to recognize it when it happens. Right? Yeah. And in that environment, what's needed? Is that robust evidence based clinical decision support, I am presented with a high risk condition, what is the best available evidence for how I should be managing this condition? And am I equipped in this lower resource Community Hospital type setting to be able to manage this patient or to think about transferring her to a higher level of care?


    Jerrod Bailey 07:32

    Interesting, okay. And we do a lot ourselves with critical access to and it's always that pattern recognition of complexity always comes up, right, the lack of the ability to have pattern recognition with complexity. So how so how does reality work? Like installs somewhere and it talks to something and it shows you something, break it down? And if you have a show and tell? I'd love to see it?


    Steven Porter 07:57

    Sure, absolutely. So I'll provide a brief overview, and then a picture's worth 1000 words. So I'll hop over to our demo environment and show you a little bit about how it works. We will commercializing this as an app as downloaded from the EMR App Store. So all the major EMR vendors, epic Cerner, medi tech, all scripts, they all have app stores now that run on this technology called Smart on fire, which is a standardized interface that third party app developers like ourselves can write to. So the way it works is we have our software as an application installed in the EMR App Store. And it effectively gets downloaded by the hospital who's interested in having it as an overlay to the EMR bill that's in place at their hospital.


    All of what you're about to see is data that is pulled from different parts of that EMR, we pulled from over 130 data feeds per patient who is run through our system. So everything from the patient's past medical and obstetric history, every vital sign from the moment of her admission for that inpatient episode of care, and the lab that's resulted any medication or blood product that has been administered. It's a whole lot of information that we mined and sift through and project on our user interface and in which we use to power our alerting, and clinical decision support algorithms.


    The biggest problem with most EMRs is what we call this aggregation of information. You have information stored in multiple different parts of the chart. And so you have this hunt and peck model where if you are trying to access the patient's lab results, you need to click over here. And if you want her recent set of vital signs, you click here but if you want to look at the 12 hour trend, it's in a different part entirely. Yeah. And when you are doing that sort of hunting and pecking, the information inevitably can get missed. What we are doing is mining for all of that critical information. We're consolidating it and bringing it together on a single user interface and we're also bringing together information shoot on both mom and baby on to that user interface? Well, when we take care of patients on labor and delivery, we're not taking care of either mom, or baby. It's a dyad. And you were trying to optimize care for both.


    So I'm going to show you our demo environment. These are some hypothetical patients. But this will give you a sense of how it works. And Jerry, you're able to see these, these five hypothetical patients, I sure can. Great. So what we're looking at here is a hypothetical labor and delivery unit, and we have five patients on our census. This is what we call the global view of risk LD. And this is accessible as a tab within the EMR. Now, what we see here is, all of the patient thumbnails have a black circle, but some of the patients have a red circle. Now the red circle corresponds to a new and unacknowledged alert that has just fired for that patient. Now, an alert could be an abnormal vital sign, it could be a new high risk condition like preeclampsia or intra amniotic infection, it could be a sign that labor is not progressing.


    Normally, there are a number of conditions that we could alert against. And we can also customize that to the hospital. So different hospitals, different units might have different alerting thresholds. But what we see here is that the patient in bed for Lisa Jones and the patient in bed six, Jane Smith, they both have new and unacknowledged alerts, they have these have just fired real time, we want to double click on those to figure out what those alerts are and to make sure that those problems are being addressed. When we look at the black circle, the black circle contains essentially, a running tally of all of the alerts that have fired for the patient sends her admission to labor and delivery eventually becomes a marker of acuity for your unit. Because if you are the charge nurse, or you are the attending physician, and you're looking at your census of patients, and you have one patient for whom two alerts have fired, and another patient for whom 37 alerts have fired, your patient with 37 Alerts is probably your higher risk patient. And you want to make sure that you have an experienced nurse at the bedside that the attending is ideally in house and involved with the plan of care that the charge nurse is aware of the plan of care. So this is essentially our air traffic control. This is how we identify patients who need intervention and allocate resources accordingly.


    So I'm going to click on this first patient Betty good case, to give you a sense of what's on her patient specific user interface. And again, this view could also be accessed from within the patient's chart in the EMR. Once this ELD is downloaded through the EMR App Store, there's a risk LD tab that would appear in the patient's chart you is the clinician working on labor and delivery clicks on the tab and this is what you see.


    Now I don't need to get overly clinical for those of your audience who haven't worked on labor and delivery. But essentially what we've done here is tried to consolidate all the relevant fetal indicators and all the relevant maternal indicators and bring them together on a single color coded user interface. So a little bit of a tour here we have the patient's name, her identifying information, her G's, and peas, which is the number of pregnancies and number of previous deliveries, we know her gestational age, and we know the estimated fetal weight of the baby. So that's all kind of on our header at the top. On the left here is a longitudinal view of the fetal heart rate baseline. This is telling us over time what's happening with the baby's heart rate since the onset of labor. This graph is showing me a visual representation of the labor curve. It's plotting two simple variables, cervical dilation and fetal station. So I have a visual snapshot of how the labor has been progressing over time.


    Now if I go over to this part of the chart, I see a couple of rows of charms, some of which are lit up. The first row of charms pertains to pre existing medical conditions. So these are conditions that the patient walks onto the labor unit. With these already in place. This particular patient has pre existing diabetes and chronic hypertension, which are conditions I want to know about if I'm managing her labor. Now, the system is also telling me that she's at high risk for what's called a shoulder dystocia, which is an obstetric complication where the head delivers and the shoulders get stuck. It's also telling me that she's at high risk of a postpartum hemorrhage, and we'll talk a moment about how we arrived at those determinations.


    This bottom row here is lighting up based on certain intrapartum conditions, things that happen or our flag while she is in labor. It's telling me her group B strep status, it's telling me whether or not she has her epidural, it's telling me the length of time since rupture of membranes. And the last bit of our tour here is basically all the parameters that we want to track when we're assessing how a laborer is progressing. We want to know the stage of labor. We Want to know if membranes are intact or ruptured? The is oxytocin on and if it's on, at what rate? Is it infusing? What is the labor progress doing? What is her cervical dilation and utilization? How is the baby tolerating labor? Have we seen the late or prolonged decelerations?


    This last section here is essentially mutter, or what we call a maternal early warning system. It's telling me Mom's blood pressure, temperature, heart rate, respiratory rate, pulse ox, and then after delivery, it is tallying her quantitative blood loss and her urine output. And the last piece that I want to show here is we actually have a scrolling toolbar that allows the witness to scroll backwards and forwards Oh, look at that. Now, this is really useful for handoffs of care we have a lot of patients who are coming in, to have labor induced for a variety of medical or other indications. Sometimes these patients are on the unit for days at a time, and there might be multiple handoffs of care. So when you're assuming care for the patient, you want to know things like, when did the patient progress from late into active labor? And when did membranes rupture? And has the Pitocin been on or off? How many times is it turned off? How many decelerations? Are we seeing? That's information that's not easily accessible in most EMRs? We've done this extracted that out? Yeah, I have a few other things I wanted to share with the alerting. But I'd stop there just to see if there are any initial questions here.


    Jerrod Bailey 16:27

    I mean, it's great. And you know, from an infant information architecture perspective, being able to dashboard all of these things, and you give a time component to this with the timeline is just, it's great. I mean, I feel like I could look at a chart like this and at least understand what I'm supposed to pay attention to. So just good from a software person to a clinician, this is this is pretty good software design so far.


    Steven Porter 16:50

    No, thank you for that feedback. You know, I think our real differentiator is that we were designed by OBS, and OB nurses for use by obs and OB nurses, which makes a difference, we really know how we want information to be packaged and presented both those of us that have worked in the trenches on labor and delivery. So it's a it's a rather simple design, but it sort of has everything that you need to focus on here as we see this bell in the upper right hand corner, which is telling me about the two alerts that are fired for this particular patient. So it's telling me that this patient has a risk factor for shoulder dystocia. And it's telling me why one because she has a history of a shoulder dystocia. During her previous delivery. She has a larger baby with an estimated fetal weight greater than 4000 grams, and she has a history of diabetes. So these are three known risk factors.


    Now, it is widely accepted that a shoulder dystocia is impossible to predict with any accuracy or to prevent should it occur. But you can mitigate your risk by counseling the patient documenting that counseling of the chart, and making sure that there's optimal preparedness on the unit that both the nurses at the bedside and the charge nurse are aware of the risk in this situation.


    We also have links to a cog, the American College of Obstetricians gynecologist, we have a cog guidance on this particular topic. So what we've extracted here is content that comes from either the practice bulletins or committee opinions that our professional body publishes on this issue. So if you're at a community hospital, and you're trying to remind yourself well, is there a is there a wait an estimated fetal weight at which I should consider a C section versus allowing this patient to labor? And you can't exactly remember what those thresholds are? We've we're displaying that content here. So it's easy to access the point of care.


    Jerrod Bailey 18:47

    Fantastic. Well, so tell me you've you've, you've been able to deploy this to places What have you learned? What kind of do you have any kind of stats or anything that you'd like to talk about? Or just kind of impact assessments of of the tool in practice? Great question.


    Steven Porter 19:02

    So we are still early stage, we went live with our second hospital, the tail end of October. What we've seen from some early looks at the data, is we're starting to see an overall narrowing in what we call the time to treatment for high risk conditions. Now, that is a process measured, but that can be a proxy for an adverse outcome. What we are looking at is the time from when a high-risk condition was flagged to the time for at which treatment was administered. We want to shorten that time. And what we've seen with preliminary analysis is that we are able to incrementally narrow that time and also reduce variation in time to treatment. From having these extreme outliers. cases that were treated hours after the condition, ideally, wouldn't be too much closer to the time of diagnosis.


    We're in the process of cause validating some more just kind of end user satisfaction with the tool. One of our biggest metrics that we want to track is, do people feel that it makes their job easier, which anecdotally people have shared that it does. Our goal is not to create any additional work, what you'll note about the interface that I was just sharing is, it's not a right back, there's no additional documentation, there's no additional workflow. It's just consolidating information that's already in the chart and making it easier to take actions against that information.


    Jerrod Bailey 20:33

    Easier. As a doctor, I'm sure it's happened, at least once you've been handed technology that makes your life more difficult, right?


    Steven Porter 20:39

    Well, exactly. And that's always, that's the hurdle that we have to overcome when we're introducing the tool is as soon as we come into the unit with a new technology, that's the fear is, oh, gosh, I'm gonna have to chart elsewhere, I have to use a different screen, I have to use a different device. But you know, I'm already overwhelmed as it is. And so we have to kind of really invest in in walking end users through actually, there's no additional work. This is how it's a time saver. Well, there's


    Jerrod Bailey 21:05

    a lot of risk and patient safety, people that listen to this podcast, whether they're carriers and TPAs. And they're trying to encourage their, their insurance or their or their clients to, to implement and think about things that are upstream from risk in order to help it downstream. Right. So there's a lot of that, certainly, they're looking for advice and ideas to bring back to their clients and their insurance to be able to help. And then we've also got folks that are sitting at hospitals or at clinics, they're, they're managing risk management, managing patient safety, what are you looking for, in terms of sort of an ideal customer? Like, what, what? And how do people reach out to you and get involved? And?


    Steven Porter 21:47

    It's a great question. I think, for this to be successful, it takes strong clinical leadership and clinical sponsorship, and also, close collaboration with the hospitals, IT department we've had situations where there might be a great clinical champion, but that clinical champion really has no sway over it decisions that are made at the hospital, and as much as they want, might want this tool, they sort of can't advocate enough to get it in the pipeline for projects to be implemented. So I think I think having that, that that sponsorship, that leadership on the clinical side, but with that close collaboration and buying on the IT side is really what sets us up for success. I will also say that the partnership with risk management is really key because risk managers, as you were saying appreciate the value of addressing upstream drivers of risk. And our position to support projects like this might even have the influence over it that clinicians do not. So it's realizing that decisions at hospitals brigade, large health systems are not made in silos, it requires collaboration across multiple departments. But where we've seen the most success is when there's, there's the interest on the part of clinician there's the buy in from risk, and together, able to influence and get that buy in from it.


    Jerrod Bailey 23:08

    Yeah, I mean, it's so important that we've seen some of the big nuclear verdicts in the last year or two years come from OB Gen as a specialty, we have clients that just send us reviews for OB GYN case, just because there's so much tied up in that there's so many things that that do go wrong, there's a lot of babies being born in a lot of places, right. And so there's just there's just a large surface area of risk there. And the idea that you can, with a single dashboard, bring that risk together and elevate for the doctor no matter what their training, what is critical to look at now is just it can't, I can't there's so much value in that. So hopefully those who are listening are thinking, hey, you know what, I'd like to suggest some solutions like this to, to my hospitals, or if I'm at a hospital to suggest this, but I just love it. I love everything about it. Well, anything else that you'd like to kind of leave us with?


    Steven Porter 24:07

    I would share, Jerrod is how useful it is for hospitals as part of their quality assurance reviews and their processes. For two reasons. One, if you are reviewing a case, and you are kind of replaying the risk LD screen, you have all the information at your fingertips that you would typically spend maybe weeks sort of leveraging chart abstractors to try to pull that story together, you have the whole story together in front of you. So it really does make those reviews much more efficient. It also really serves as a great teaching tool, if you have new orienting nurses or medical students. Gosh, yeah, it's the whole story together.


    The last piece that we're able to do, which I'll show you some quick screenshots of is the reports that we're able to issue with our software. Now what we realized I'll scroll to the top here is that because we collect so much data from the EMR to power our alerting and decision support and algorithms, were in a position to analyze that data and issue real time reports back to the hospitals about the use of risk. So we can tell them in real time, how frequently are you encountering some of these high risk conditions? And how effectively are you managing them. So this is a hypothetical report. This is all hypothetical information, but it's based on actual reports that we do issued for our clients.


    And I'll just give you a quick tour here. So we're showing just how the delivery volumes are flexing up and down month to month, we're showing the frequency with which they're encountering certain high risk conditions, the different kinds of hypertensive disorders of pregnancy, the cases of shoulder dystocia, or OB hemorrhage, intra amniotic infection and diabetes, we interesting story, one of our hospitals, they realized, after a couple of months of looking at this dashboard, that their rates of diabetes on their eat, it was much higher than they thought it was. And they realized they needed to implement some additional education and training on managing diabetes in a laboring patient, because it's very different from managing diabetes in the office. But they didn't have that visibility before our dashboard of how we have a significant percentage of patients on the floor that have either pre gestational or gestational diabetes, were able to capture the timeliness of nursing documentation. This is a real pain point for some hospitals, particularly hospitals, as a lot are now that have traveling nurses that might not be aware of hospital protocols or policies for how frequently to take vital signs or assess pain or interpret the fetal heart rate strip, we can feed back to the unit how, how compliant. Were your staff with documenting on time versus what percentage of the time was that documentation delayed? And are there opportunities for improvement.


    And this last section, that all that I'll direct you towards here is this time to treatment section. So we talked about tracking, this is a process metric. But of all of this, to give you an example, with a case like intra amniotic infection, which is an infection that can develop within the uterus during labor, we can tell you of all the cases that you had in the last month, how many got treated within the hour, which is the ACOG standard, versus how many had delays in treatment of between one and two hours or greater than two hours or whether in cases for which there was no treatment. And so we do this right now for three conditions, we do it for intra amniotic infection. We do it for severe preeclampsia measuring the time from that diagnosis to the administration of magnesium sulfate. And we also do it for treatment of severe hypertension. And what we do is hand these reports back to the client and Mrs. Jones for cases where there was either no treatment or significant delays in treatment will get reviewed. And ultimately, it's the termination of the hospital of their l&d or their quality of leadership.


    Do we mean our standard of care here? We're not necessarily accusing anyone of practicing that medicine, we're just saying, hey, our system flagged that there was a more than two hour delay in administering the antihypertensive? Why was that? And the teams can blue look into why that's the other piece that's valuable. You know, if there was a significant delay in administration of antibiotics was that because we didn't recognize the condition, we didn't place the order, the order wasn't validated, the medication didn't come up from pharmacy at what point in that chain did did things break down? If you don't know the root of the problem, it's hard to kind of implement your solution. What this does is really focus those teams on cases, which may not have resulted in adverse outcome, maybe it was just a near miss. But we can say, You know what, there's a delay in care here. And you might want to look at what was driving that? And can we implement systems processes to prevent that from happening.


    Jerrod Bailey 28:49

    And that is consistently the best philosophy that I see, as we as we work with we ourselves work at different hospitals and clinics, the ones who are looking for the things that don't, that haven't ended up in a unintended outcome, the ones that are looking ahead of that the ones that are playing proactive, this we all have limited resources, what you're doing is you're helping elevate the things that can be looked at to find the opportunities to make to make delivery better, right. And so I just love everything about this being able to help play ahead, rather than reactive. And if things get punitive and reactive later down the stream, right, when these things are becoming lawsuits and other stuff. I think our goal in healthcare should be being proactive.


    Steven Porter 29:34

    Exactly, exactly. You know, what I'll share with you Jerrod is that we've actually have some clients that have, have thought about leading with this report. And what we can do with our software is actually kind of install it in the background so that you don't have any end users that are working with it. It's not actually alerting in real time on the unit. We're just running these reports on a month. It's fantastic data, and we use that to identify areas of risk and you might find out After a couple of months we haven't missed a single case of severe preeclampsia in three months, we are really good. And missing that. But you know what, we had a couple cases of OB sepsis that we didn't pick up because sepsis presents differently in the etc. And so using the results of this reporting tool to then design a solution that maps to the areas of risk. So


    Jerrod Bailey 30:25

    fantastic. I love it. Well, Dr. Porter, this was great. How do people find out more? Where should we send them? Obviously, we'll put links in the show notes here. But where should they go?


    Steven Porter 30:38

    Absolutely. So I would direct you to our website riskLD.net has all our contact information on there, folks are welcome to contact me directly sporter@riskLD.net. And, and we'd be delighted to hear from anyone that's interested in in a demo or learning more about what we do.


    Jerrod Bailey 30:55

    Fantastic. Well, thanks for the work that you do. Love, love anybody that's using good technology to solve some real problems. Well, for everybody else, thanks for listening to reimagining healthcare, a new dialogue in risk and patient safety. Subscribe and Share if you found it valuable. And if you'd like to participate as a guest again, just email us at speakers at medplace.com. And be sure to follow Steve. Again links that will be down below. And Steve, thanks so much. This is great. My pleasure. Thanks for having me. It's good to see you again. Great. All right. We'll talk soon. Thanks, Jerrod

Experienced OB hospitalist and CEO of riskLD, Dr. Steve Porter, MD explains the riskLD approach to maternal care. He shares how siloed data can create risk during pregnancy, and how to use technology to ensure the data stays readable and actionable. Dr. Porter demonstrates the riskLD platform, and explores how to collaborate with hospitals to build the tools they need to deliver excellent care. 


Guest - Steve Porter, MD

OB Hospitalist and Chief Executive Officer at riskLD

Dr. Steve Porter, OB/GYN, is passionate about bettering clinical outcomes on a global basis. His residency was under the tutelage of Dr, Nancy Cossler at University Hospitals, and he operationalized that knowledge by refining and iterating the riskLD platform as a one day-per-week hospitalist at UH since inception. Dr. Porter holds a BA, summa cum laude, from Princeton (recipient of the Moses Taylor Pyne Honor Prize), MBA from Harvard Business School, MD with honors from Harvard Medical School and helped launch Zikani Therapeutics during his Blavatnik Fellow in Life Sciences Entrepreneurship at HBS.

View More Related Content

Ready To Get Started?

Whether you're ready to request a review or want to see the Medplace platform, we're available to help.