Nadia Huancahuari, MD: [00:32] Before I start, I wanted to just mention something. We wanted to share this data to give you a sense of how this process progresses over time, right? But I think you both have brought up the important point of, how do you address a case, like one particular event? And how do you gather that demographic data to address that one particular case? And then how do you look at it from a macro view when you're looking at other metrics such as CLABSI or, or CODIS and then you can just do more general demographic breakdown. But, um, I get a sense that, you might be feeling that you're going to get a ton of cases and it's going to be really difficult to do this manual extraction, if that's how you're going to start, right? And of course, automated is always, it's always the best way to do it, but hopefully the next few slides will give you an insight into how our process started, and you'll get a better sense that this is doable, even if you had to do it manually.
[01:34] Okay, so this slide here shows the total number of cases from 2020, 2021, and 2022. So as you can see, for quarter one of 2020, there were no cases that were flagged for equity, because the actual work in our tracker began in May of 2020. So you can see that at the beginning, despite how much you may socialize this, you are depending on frontline clinicians to report the safety reports, right, and to flag it for equity. You're also depending on whoever has interfaced with patients. So in our case, patient-family relations specialists to receive these complaints for patients and flag it for equity. So look at the beginning, in that second quarter of 2020, there are only 17 cases. And just to give you a sense of our institution in the number of safety reports that are applied per year, or that the we obtain per year, that we get per year—it varies between 12 000 to 14 000 safety reports filed at our institution. So look at the numbers that you're seeing right now, right? So it's going to take a bit of time to get to those high levels. Maybe yours will be different and your journey may be a little bit different, but this is our journey. So second quarter 17 cases, and the last quarter of 2021, one of our highest number of cases, 45. And right now, this isn't the full quarter for the, for 2022, but we already had 24 cases.
[03:15] I wanted to just point out a couple of things that happened here that you may see when you go through this yourself. Two things started happening as we continued to socialize this, this concept of looking at events from an equity lens. One was that although frontline clinicians knew to flag for equity concerns, sometimes the cases were not being flagged for equity concerns, and then when they would go for review by their risk managers or local leadership, then they now knew to flag it. So then they would go back and flag it and eventually it would make it to us. So it was it was really interesting to see that little bit of a shift in culture that other people were recognizing it. So you were no longer depending just on frontline clinicians but you—or PFR specialists...or patients—but also another layer of people that were going back and flagging these cases. So that was important to see.
Huancahuari: [04:21] The other thing that will contribute to an increased number of cases that you're going to receive is the reputation of your team. So I'm very fortunate to work with a team that is really invested in this. So as the safety reports come through, we try to respond to them within the next few days, but definitely within 7 days. And so then, the institution or the enterprise ends up recognizing that there's a team that is looking at these cases very closely because you're connecting with them. And I think one of the differences—as you can imagine based on the number of safety reports that we receive—in general we do not connect with those reporters to let them know what happened to their report, right? So most people are filing reports, and they were never going to hear about what happened with that safety report. But because the numbers are manageable, at least for now, we make contact with these reporters, so they're constantly hearing that we are reviewing these cases. So...and this goes back to our connections with risk managers, patient safety specialists, and again, departmental leadership. So that will also help further socialize this effort and give you an increased number of cases. Next slide, please.
[5:41] So as you work through these cases, you're going to have... As you're doing the work, there'll be action items per each case. So this next graph here that you're seeing shows that work. So again, based on the number of cases, which I would say like 30 to 40 or so, these are the number of action items. And these...and what we mean by action items is any work that happens on the case, which may range from a request for clinical reviews by departmental leadership, specific educational efforts, family meetings with patients to address some of these concerns after the review has occurred. We have specific process improvement initiatives that come out of these, these reviews. Also referrals to specific committees. So for instance, we have a committee which is patient at-risk committee, which manages cases where there was discrimination from patients towards providers. And we will refer those cases as they come through to that committee. But again, so these are all...this work constitutes action items at various levels, right? Like systems levels, behavioral levels. And so once you do all of that work—-next slide, please—you're able to have something like this. So, and this is an example this is not specific to our institution. But you can then look at all of your actions and see where these actions fall. And so you may have...like a bigger part of this may be system contributors, or maybe for you, it's performance contributors. And we taught it at Brigham, we have not been fully focused or our effort didn't start with social contributors and structural contributors, but that is something that we definitely want to pursue further. And this is how you've been...but you are being trained on all of these through this peer network. So we wanted to include that in the pie chart. Next slide, please.
[07:49] We wanted to share that 2021 data and regards to themes. So as you're doing this work, you're going to see repeated themes coming across as you're reviewing these cases. And these are not in any particular order. These are sort of like our core themes, our most common themes that we wanted to share with you. So racism, explicit or implicit, will come out. Lots of issues regarding language barriers. So there's a, there...you're going to find issues about that have to do with frequency, how frequent they are, and also some of them that are more impactful. SOGI discrimination was something that we've worked on, too, we had a number of cases, we're really grateful to those patients who raised these concerns. And like anything in safety and inequality, and anything with adverse events, communication and issues of communication, it's a really prevalent theme in our tracker. Access to care came up a lot. And that had to do with, not just specifically access to care, but in combination with some of these other themes. So for instance, patients who were unable to access care due to language barriers, like they couldn't schedule the appointment, or patients whose insurance wasn't being accepted by that particular community clinic. And a lot of the times, it wasn't the fact that the community clinic wasn't accepting the insurance, it was that there had to be further education, and with the administrative staff to recognize this. So lots of opportunities for improvement from that perspective. And the other one, which may be specifically related to COVID is visitation policy and enforcement. So we've had a number of issues related to that. And we'll see that persist as the years progress.
Huancahuari: [09:38] But one thing I wanted to highlight here is that as you're seeing these themes emerge, you're going to feel the need to address them all and you can't, right? And so you have to think about frequency and impact. And so together with your team, you'll decide what is something that may be easier to accomplish? It might not have like a super high impact, but maybe it's cases that are happening again and again and again, and it's worth addressing it. And you'll find things that you know, you want to definitely fix or improve, but it's going to take many years. Say something that is related to like, the actual, like Epic, or whatever your medical record is, and that particular intervention will take many years to accomplish. It's something that it's going to be on a list of things to do, but something that you might not be able to address right away. But this type of breakdown will allow you to figure out where you can invest your energy and your limited resources.
[10:43] Last thing I'll say is that, as you're going through these themes, you'll have opportunity for process improvements. And again, you'll work with others and partner with other departments in leadership to support that work. And so the next slide will let Esteban talk about a particular QI initiative related to breastfeeding.
Esteban Gershanik, MD: [11:08] Thanks, Nadia. And just as you highlight it, you know, it's a matter of a lot of these themes, once you start capturing the numbers and knowing what it is, you can then sort of start breaking down the data. And so here's just one example about a lot of the safety issues we're hearing and then how with our OB group and our NICU group and maternal services, how they integrated basic data, equity data to initiate a quality initiative improvement with breastfeeding. And so one of those measures that we have is around breastfeeding rates at discharge. And when you know, and looking at this, we just simply broke it down between white, Black, English speaking and non-English speaking, and saw such a great difference between all of these. And one of the things that also comes to my mind is to quote my colleague, Reagan, you don't want... You know, we were discussing earlier about what's all the data pieces that we need, and as Karthik mentioned, as well and what Reagan would always say is like, don't let the progression of breaking down this data, don't let perfection get in the way of progression. And so just looking at this, you can dramatically see a difference between different races and language initiatives when it came to breastfeeding rates at discharge. And one of the things that we further asked was the aspect of like, well, were they intending to breastfeed for Hispanic mothers? And were they at discharge? And seeing that there's a large gap between intent and what we were doing, and one of the things that this led to was eventually having us understanding that many of our lactation specialists were white women that only spoke English. And so one of the things that this led to from a quality initiative and action standpoint was in hiring lactation specialists that were of different races and spoke different languages. And we're already seeing QI projects on this improve some of these efforts. If you want to go to the next slide.
Karthik Sivashanker, MD: [12:52] Oh, and let me add one thing on that last slide, Esteban, that was, that was a good example. So what I want to add for y'all is, first, this is an example of the type of quality improvement work that can and should emerge from your equity informed high reliability work around harm event. And what we're really saying here is, don't try to develop a large-scale QI initiative for every single case and every single contributing factor, because you'll quickly get paralyzed. Rather, start to notice what themes are emerging—and we're actually going to share in the coming months, a more standardized taxonomy for that—but for now, it's really just, we're noticing language over and over again. Start tracking those themes, and then be very intentional around which opportunities you're going to pursue in this way. And I'm going to link this back up to what I said earlier about race, for example, being a social, not biologic construct. This is where it's really important to keep reminding yourselves, but also the folks around you who are not as familiar with health equity data. This is not saying that, for example, Black women don't want to breastfeed because it's some biologic difference in them, right? This is saying that there are differences in breastfeeding rates that are associated with how these patients are self-identifying, and that probably has to do with a lot of factors, including things like bias and discrimination and so on. So we want to close that gap or eliminate that gap to the extent possible, while also respecting the patient and community preferences. So go ahead, Esteban. I just wanted to throw that in there.
Gershanik: [14:38] Yeah, no, thanks so much Karthik for adding that in. You know, I also think of Karen Fiumara, she's always saying to maintain your innate curiosity on this. So to see that there was such differences around lower rates among Black and Hispanic women than white, and lowest rates of exclusive breastfeeding amongst Spanish speaking Hispanic women, and knowing that in my global health experience, you know, one of the breastfeeding clinics that we had, or one of the pediatric clinics we had in Honduras, they were 100% breastfeeding. So it's amazing how from one country to another, you can have such differences. And to maintain that curiosity to understand like, what are the reasons for this? And then addressing it further on.
[15:12] But just to go to the next slide, in the interest of time—because I want to make sure we have some conversation on this—is while this was what things were like within our hospital walls, one of the things that we started was, again, hiring inpatient wise, and starting this program called Mama Sana for those healthy mothers for those who don't know the translation from Spanish to English. And we started having the initiative with our Spanish speaking lactation specialists while they're inpatient, but also understanding that our patients or our people in the community aren't just in the hospital, they live outside the hospital, and making sure that we line things up between what partnerships we have in the outpatient basis—so knowing that our outpatient clinics had some family partnerships as well—and that our communities also had several strong resources. And in this scenario, Stronger Generations Baby Care in Boston Breastfeeding Village, they all address breastfeeding not just for people that are part of our network, but people that are in the community in general. And so lining these things up and ensuring that their patients centered along with equitable, were resources that lined up the journey for these patients and made an impact. Go to the next slide.
[16:01] So in getting some of this data, one of the things that I think we've learned at Brigham is we started asking some basic questions in collecting some of our data. In a nutshell, what is the demographics of the patients we serve? Well, that sounds like a crazy question. Many of us don't know that off the bat. And the idea of who comes through your emergency department? Who's admitted to your hospital inpatient basis? What service do they go to? Where are they being discharged to, whether it's a SNF, an LTAC? And who is seen in your ambulatory practices? Not just primary care, but everything from your cardiology, dermatology, neurosurgery, and understanding and comparing that to the baseline demographics of the community that you're, that's around you, and starting to get some of that narrative description and quality documentation for each of these cases, and engage in both patients in the community and understanding and making a difference in who you serve.
Sivashanker: [17:10] I do want to encourage you all to get that baseline demographic information of your patients, which is who are you currently serving? And then the baseline demographics of your community, or who's in your catchment area. And we can also provide you some guidance on how to define that, which is really saying, who should you be serving? And ideally, we should see is that who we are serving matches who we should be serving, ie, that we're taking the care of a fair share of our patients. Oftentimes, that's not the case. And I'm just going to be transparent, because that's how this work is going to be impactful is that, you know, at more affluent academic medical centers... And so that's going to be important context as these events start to emerge. Because if you notice, for example, access issues emerging, well, it's important to note, be able to link that... Well, we're having these access issues for, for example, our Medicaid patients, and we're chronically under serving our Medicaid patients. So now you're starting to link it to the institutional level data.
[18:12] What I want to really encourage you all to think about is starting low tech. So I'm going to give the story of the Brigham and Women's as an example of how starting high tech is not always a great idea. So at the Brigham, before I actually joined in my role, our DE&I and community health teams had spent a good number of years actually developing a very sophisticated dashboard, a health equity dashboard. And that happened right around the time we transitioned to Epic, and by the time all of that was said and done, nobody was looking at the dashboard or using the dashboard. And it was a lot of work that ended up not being used in the service of patients like it should have been. So that's not to say that dashboards like that aren't valuable. It's just don't put that in front of the work that needs to be done. So start by just keeping it simple, manually extract your data, the demographic and other data, start integrating it into your trackers, and get that retrospective data. That's going to let you do a pre/post comparison. And all of these are things we've asked you to track on your checklist because at the end of this whole initiative, we want to be able to say, for example, before we started this work, here's how many equity events, equity harm events we had, and after we started it, here's how many we identified. Before we started and here's how many actions we were taking around these events, and after here's how many actions. And then we want to be able to describe how many quality improvement larger initiatives have been initiated out of this work, and being able to describe qualitatively as well as quantitatively the outcomes of that. So low tech, start now, manual, and dashboards are a nice-to-have but not a must-have.
Credit Designation Statement: The American Medical Association designates this enduring material activity for a maximum of 0.25 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
CME Disclosure Statement: Unless noted, all individuals in control of content reported no relevant financial relationships.
If applicable, all relevant financial relationships have been mitigated.