00;00;07;20 - 00;00;09;28 Music Track [music] Tommy, can you hear me? [music] 00;00;12;13 - 00;00;39;06 Ken Milne Welcome to the Skeptics Guide to Emergency Medicine. Meet em, greet em, treat em, and street em. Today's date is November 28, 2022. And I'm your skeptical host, Ken Milne. The title of today's podcast is Tommy Can You Hear Me? Deaf and Hard of Hearing Patients in the Emergency Department. And our guest skeptic is Doctor Corey Heitz. He's an emergency physician at Roanoke, Virginia, and he's also the CME editor for Academic Emergency Medicine. 00;00;39;22 - 00;00;48;14 Ken Milne Welcome back. To the SGEM, my friend. And since you were on the show last time, you've had a big life event. Want to share? 00;00;49;25 - 00;01;08;29 Corey Heitz Yeah, Ken. So, I got married September 24th. I married my wonderful now wife and we went down to Asheville, North Carolina, then two weeks in Greece, and it was amazing. I've it's only been a couple of times that I've been on a two week trip and two weeks is a great time to be gone. You really get to do a lot of things. 00;01;08;29 - 00;01;26;06 Ken Milne And well, I followed you and all your adventures through Greece remotely on Facebook, and it was it was great. You look happy. Congratulations. Wonderful news. But let's get started because this is a the jam, not the Corey Jam. So give us a case. 00;01;27;01 - 00;01;46;22 Corey Heitz All right. So one night you're working in the emergency department and you grab the next patient on the board and upon entering the room, introduce yourself as normal. The patient waves at you and gestures to a friend in the room who explains that the patient is deaf and needs a sign language interpreter. You know, from your brief look at the triage report, that the patient's vital signs are stable and their chief complaint isn't likely to be immediately life threatening. 00;01;46;22 - 00;01;51;02 Corey Heitz So you politely explain, you'll be back and go find the charge nurse to obtain interpreter services. 00;01;52;19 - 00;02;10;21 Ken Milne The emergency department sees anyone, at any time, for anything. This includes patients who have difficulty accessing the health care system, uhh, so people with certain socioeconomic status, race, gender, mental health, substance use disorder, and physical disabilities. 00;02;12;12 - 00;02;22;27 Corey Heitz Deaf and hard of hearing patients experience disparities and social outcomes as well as health inequities likely due to audism, which creates privilege for non deaf and hard of hearing people in our society. 00;02;23;22 - 00;02;51;00 Ken Milne DHH, or deaf and hard of hearing, patients are more likely to use the emergency department than non DHH patients. But little research has been done to compare ED focused outcomes for these groups of patients. DHH patients are heterogeneous with adult onset DHH patients being less likely to use American Sign Language or ASL with proficiency. 00;02;51;24 - 00;02;59;08 Corey Heitz DHH ASL users may have delays due to interpreter availability, potentially resulting in care discrepancies. 00;02;59;08 - 00;03;05;04 Ken Milne All right. So that's the background information. Corey, why don't you give us the clinical question we're going to try to answer on today's podcast? 00;03;06;02 - 00;03;18;08 Corey Heitz How do deaf and hard of hearing American Sign Language speakers and DHH English speakers utilize the ED? Specifically, regarding acuity of complaints and pain? What is their length of stay and what is the prevalence of acute revisits? 00;03;19;06 - 00;03;19;28 Ken Milne And the reference? 00;03;20;20 - 00;03;31;23 Corey Heitz James et al. Emergency department condition acuity, length of stay, and revisits among deaf and hard of hearing patients: A retrospective chart review from Academic Emergency Medicine, November 2022. 00;03;32;13 - 00;03;42;23 Ken Milne Oh, that means this is HOT OFF THE PRESS. Yes! Another SGEMHOP. This is the November issue. So, let's go through the PICOT. What was the population? 00;03;43;16 - 00;03;51;23 Corey Heitz All DHH American Sign Language, DHH English speakers, and non DHH English speakers who had used a single academic emergency center for care. 00;03;52;18 - 00;03;53;24 Ken Milne And who did they exclude? 00;03;54;22 - 00;03;57;15 Corey Heitz Patients who had not had an ED visit during the time period. 00;03;58;12 - 00;04;04;21 Ken Milne And there wasn't really an intervention because this was looking at how these people were accessing care. What were their comparisons? 00;04;05;24 - 00;04;11;17 Corey Heitz Non DHH English speakers were compared to DHH ASL users and DHH English speakers. 00;04;12;14 - 00;04;15;13 Ken Milne All right, let's go through their outcomes. What was their primary outcome? 00;04;16;08 - 00;04;27;00 Corey Heitz They used several different primary outcomes. The Emergency Severity Index, the ESI, triage pain score, ED length of stay, and acute ED revisits defined as within nine days. 00;04;27;28 - 00;04;29;29 Ken Milne And what type of study are we looking at today? 00;04;30;11 - 00;04;33;00 Corey Heitz This was a retrospective chart review of a single health care system. 00;04;34;00 - 00;05;00;04 Ken Milne Well, I did spill the beans that this is an SGEMHOP. So that means we have the lead author on the show, Doctor Tyler James. Tyler is a post doctorate research fellow in the Department of Family Medicine at the University of Michigan Medical School. His research focuses on health care access, utilization, and delivery for people with disabilities, and specific interest in working with people with sensory disabilities. 00;05;00;22 - 00;05;14;29 Ken Milne He is also a mixed method research methodologist, and serves the associate editor for the Media Review of the Journal of Mixed Methods Research. Welcome to the SGEM, Tyler. 00;05;14;29 - 00;05;15;22 Tyler James Hi there, glad to be here. 00;05;17;07 - 00;05;23;09 Ken Milne You know, is there any particular reason that you got interested in this area of research? 00;05;24;01 - 00;05;46;04 Tyler James So sure thing. The short answer is yes. The longer answer, though, is that both of my parents are deaf and hard of hearing, but they both use spoken English. My mom, since she was three due to an illness in her childhood; my dad had occupational exposure in his twenties, in the military. And so, I - growing up with deaf and hard of hearing parents, I had a very different experience than most children. 00;05;46;04 - 00;06;10;06 Tyler James I think making sure that we had face to face communication and discussing things in person that rather than across the room or like yelling across the house. My cousin was also culturally deaf, which means that she uses American Sign Language. She's a part of the linguistic and cultural minority group of deaf people who use American Sign Language. And growing up, Mom and I had always said, "Oh, we would learn sign language later." 00;06;10;06 - 00;06;45;18 Tyler James In college, I had the opportunity to learn sign language at around the same time that I was entering my PhD program. But I had no interest in working with people with disabilities. I was focused more on HIV related stigma and how that impacted testing utilization and treatment. I was at a lunch with a deaf friend and mentor who is my mom's age, and he thought he was having his second heart attack. And he had a wealth of comorbidities, including poorly managed diabetes, congestive heart failure, just - just a wealth of comorbidities. And so we ended up calling an ambulance and going to the hospital. 00;06;46;01 - 00;07;07;20 Tyler James And in the community we had heard stories about, "Oh, it takes forever for an interpreter to arrive to the emergency department." And in our area, especially because we were in a more urban area surrounded by a huge rural area. So it wasn't a large metropolitan area. And going to the ED, I advocated for over 8 hours before an interpreter arrived. 00;07;07;29 - 00;07;25;18 Tyler James And so although he was not having a heart attack, I just imagined, well, there could be a wealth of inequities coming up, both in ED care and also with an inpatient care. And so around that time, I also met my current post-doctoral mentor, Dr. Mike McKee, a deaf family medicine physician who encouraged me to go down this route of research. 00;07;25;29 - 00;07;27;24 Tyler James And so that kind of led me to where I am now. 00;07;28;20 - 00;07;51;17 Ken Milne I love hearing stories like that. You know, people get into research for individual reasons, and that is one of the best stories I've heard recently about how someone ended up in their area of research. So thanks for sharing. We're going to start this off with your participation and just ask you, what did you and your coauthors conclude in your paper? 00;07;52;16 - 00;08;14;22 Tyler James Yeah. So our study identified that deaf and hard of hearing American Sign Language users have longer emergency department length of stay than non deaf and hard of hearing English speakers. Although additional research is needed to further explain that association between deaf and hard of hearing status and ED care outcomes, including length of stay and acute revisit, which may be used to identify more intervention targets to improve health equity for this population. 00;08;16;15 - 00;08;26;09 Ken Milne All right, Tyler, Corey and I are just going to run through the checklist and the key results, and then we're going to bring back you into the conversation and ask you five nerdy questions. So get ready. 00;08;26;20 - 00;08;27;05 Tyler James Sounds good. 00;08;27;21 - 00;08;34;08 Ken Milne Corey. Here we go. Checklist observational studies. Did this study ask a clearly focused issue? 00;08;35;03 - 00;08;35;19 Corey Heitz Yes, it did. 00;08;36;15 - 00;08;39;23 Ken Milne Did the authors use the appropriate method to answer their question? 00;08;40;16 - 00;08;40;27 Corey Heitz Yes. 00;08;41;11 - 00;08;44;05 Ken Milne Do you think the cohort was recruited in an acceptable way? 00;08;44;25 - 00;08;49;02 Corey Heitz I think so. I think we're going to discuss the details of that a little later. 00;08;50;01 - 00;08;52;11 Ken Milne Was the exposure accurately measured to minimize bias? 00;08;52;11 - 00;08;54;11 Corey Heitz Yes. 00;08;54;11 - 00;08;55;05 Ken Milne How about the outcome? 00;08;56;11 - 00;08;56;20 Corey Heitz Yes. 00;08;57;11 - 00;09;01;00 Ken Milne And the authors, have they identified all important confounding factors? 00;09;01;19 - 00;09;02;07 Corey Heitz It appears so. 00;09;02;28 - 00;09;05;00 Ken Milne Do you think the follow up of subjects was complete enough? 00;09;05;19 - 00;09;06;00 Corey Heitz Yes. 00;09;06;23 - 00;09;07;29 Ken Milne How precise are the results? 00;09;08;26 - 00;09;09;16 Corey Heitz Fairly precise. 00;09;10;03 - 00;09;13;01 Ken Milne Do you, Corey, believe the results? 00;09;13;20 - 00;09;14;05 Corey Heitz I do. 00;09;15;04 - 00;09;17;24 Ken Milne And do you think you can apply it to your local population? 00;09;18;13 - 00;09;20;24 Corey Heitz That's a little harder to say. I've got to say, unsure for that one. 00;09;21;17 - 00;09;24;28 Ken Milne And do the results of this study fit with other available evidence? 00;09;26;00 - 00;09;26;11 Corey Heitz Yes. 00;09;27;01 - 00;09;30;12 Ken Milne And then the final question, what about funding for the study? 00;09;31;13 - 00;09;42;25 Corey Heitz The study was funded by the Agency for Health Care Research and Quality and the National Center for Advancing Translational Sciences of the National Institutes of Health under a University of Florida Clinical and Translational Science grant. 00;09;42;25 - 00;10;14;28 Ken Milne Alright. Let's run through the results. The sample included 100% of the deaf or hard of hearing American Sign Language people and so that was 277 people. And this was compared to 1000 random sampled, deaf and hard of hearing English speakers and 1000 randomly sampled, non deaf and hard of hearing English speakers. Of this total sample, about a third in all three groups had an ED visit during the study time frame that was analyzed. 00;10;15;11 - 00;10;25;07 Ken Milne The mean age was in the mid to late forties and just over half were women and two thirds were identified as white. What was the key result, Corey? 00;10;26;09 - 00;10;36;21 Corey Heitz There were no statistical differences in emergency severity index, triage pain score, or acute ED visits. But DHH ASL patients had longer ED length of stay. 00;10;37;16 - 00;10;44;03 Ken Milne All right, let's dig into those primary outcomes then - outcomes...emergency severity index, give us some more details. 00;10;45;00 - 00;10;55;07 Corey Heitz When compared to non DHH English speakers, neither DHH ASL users nor DHH English speakers had higher odds of being classified into lower acuity ESI levels. 00;10;55;29 - 00;11;00;27 Ken Milne All right. So that was the emergency severity index. How about their triage pain score? Did they differ. 00;11;01;27 - 00;11;11;10 Corey Heitz On a scale of zero to ten, the mean score was 5.8 and the median was seven. Neither of the patient groups had pain scale rating significantly different than non DHH English speakers. 00;11;11;22 - 00;11;19;00 Ken Milne So they're coming in and registering with the same severity and the same amount of pain. How about the acute ED revisit? 00;11;20;08 - 00;11;28;25 Corey Heitz This was defined as a return within nine days. 10% of patients had acute revisits to the E.D. There was no statistical difference between the groups for this metric. 00;11;29;10 - 00;11;37;13 Ken Milne So some people refer to that as a bounce back rate and there really wasn't a difference in the bounce back rate. How about length of stay in the emergency department? 00;11;38;14 - 00;11;53;12 Corey Heitz DHH ASL using patients stayed in the ED 9% longer than non DHH English speaking patients. This equated to about 30 minutes longer in the ED. There were no significant differences between the DHH English speaking patients and non DHH English speakers. 00;11;56;24 - 00;12;12;16 Ken Milne All right, so that covers mainly the result section. Now it's time to do what I love to do. Let's talk nerdy. All right, so we're going to ask Tyler five nerdy questions and ask him to respond to each one. And I'm not going first this time. Corey, you get to go first. 00;12;13;25 - 00;12;29;25 Corey Heitz OK. So, Tyler, selecting the cohort, can you explain the decision to start with the cohort of patients who utilized any of the medical center facilities and then select those who presented to the ED as opposed to just isolating the patients from the ED visits themselves? 00;12;30;14 - 00;13;09;24 Tyler James That's an excellent question. So this was related more to my dissertation's first and second aims, which were related to ED utilization during the past 12 months and 36 months. And so for that, we needed patients that both used the ED and did not use the ED from the entire health system. And so that paper was a replication and extension of an article from back in 2015 that was conducted at the University of Rochester Medical Center, which you or the listeners may or may not know that has the largest per capita deaf and hard of hearing American Sign Language user population, including deaf scientists and medical doctors that work at University of Rochester Medical Center. 00;13;09;24 - 00;13;33;14 Tyler James And a very large university, the National Technical Institute for the Deaf There, the Rochester Institute for Technology. And so I hypothesized that my local setting in Gainesville, Florida, was going to have different demographics and different health care utilization patterns. Just on the differences in access available both in Rochester and in Gainesville. And so in order to replicate and extend that study, we first needed to have a cohort that had both users and non users. 00;13;34;00 - 00;13;38;28 Tyler James So that cohort then led into the cohort that was used for this paper for these care outcomes. 00;13;39;28 - 00;13;51;08 Corey Heitz Gotcha. So that makes sense. So you're saying you basically had already - So this is this - this research is part of a larger body to be in your dissertation? [Tyler: Yeah.] You already had the data set and so you used that to then? 00;13;51;16 - 00;13;52;14 Tyler James That's correct. 00;13;52;14 - 00;13;52;23 Corey Heitz Write this paper. 00;13;52;23 - 00;14;19;00 Tyler James Yes. So I think in total my dissertation, I should know this off the top of my head, I think it was six primary aims and then the revisit was an exploratory aim for the quantitative study. And this was a part of a larger mixed method study where I was synthesizing the available literature to develop a conceptual framework of both patient and non patient factors related to ED use and ED care for deaf and hard of hearing patients. And then testing some of those models, assumptions and variables within that model, both in the quantitative aim 00;14;19;00 - 00;14;20;23 Tyler James and then I had a follow up qualitative study. 00;14;21;17 - 00;14;35;27 Corey Heitz Yeah, That makes sense when you're looking at all all medical center users: who goes to the ED? When an ED researcher does a study to see who used their ED and how they used it, they usually start with patients who came to the ED, right? So kind of came at it from a different tactic. 00;14;38;09 - 00;15;01;02 Ken Milne So the second nerdy question is about this return visit. Often in the ED literature, we look at a 72 hour, a three day bounce back rate or even a seven day bounce back rate. Sometimes we'll even go out to two weeks or even one month. Nine. Nine seems like a weird number. So why did you decide to use nine days for your acute ED visit metric? 00;15;02;06 - 00;15;27;10 Tyler James Absolutely. And this is a great question, and I'm so excited that it came up on the list of questions to be asked because this was something my committee grilled me on. And so the Agency for Health Care, Research and Quality, they have a diagnostic framework or guideline related to ED discharge failure, which is not receiving a high quality discharge where patients feel that they know what their follow up plan is or feel supported and have the resources necessary to implement their treatment plan. 00;15;27;25 - 00;16;11;14 Tyler James And so acute revisit is one of the markers of diagnostic or not excuse me, not diagnostic, but is one of the markers of discharge failure. Now, AHRQ lists exactly what you just listed. 72 hours, up to 30 days of this range of a bounceback. But, back in 2014, ED health services researcher Kristin Rising and colleagues published an article in Academic Emergency Medicine where they looked at a time to event analysis. Specifically, looking at the optimal time to revisit cut point and based off of their literature review they identified that ACEPs guidelines in addition to what was being reported in the literature had no empirical backing. And so they develop time to event model using 00;16;11;14 - 00;16;33;06 Tyler James data from Nebraska and Florida to look at all of their ED care and what specifically was the optimal cut point. And through that analysis, they identified two distinct populations: early returns and late returns. Early returns, they found, uhh- were typically related to the index ED visit condition, whereas late returns were about something different or more about chronic health issues. 00;16;33;17 - 00;16;54;10 Tyler James Among the early returners, though, 99% of them were captured within nine days. And so they found that nine days was the most optimal cut point for this. And so late returns again, were not necessarily -- we didn't look at revisits for similar issues, although we do have the presenting concern. So that is something we could look at. But typically we would expect that late returns, so nine days to 30 days, 00;16;54;10 - 00;16;57;13 Tyler James would not be returning for something similar to their index ED visit. 00;16;59;06 - 00;17;17;03 Ken Milne I love that explanation. That is great, you know, because when you look at it, you know, it seems somewhat arbitrary. And I love - I love that there's data behind it because, you know, Corey knows that five is my favorite number just because I can count to it on one hand. And of course we do the metric system, well in most of the world 00;17;17;03 - 00;17;44;26 Ken Milne we do the metric system come on, America [laughter] system. And then of course seven I mean in 321 A.D., Emperor Constantine codified that, you know, a week would be determined to be seven days. So now we're prescribing everything in Constantine units, one Constantine units represents seven days. Two Constantine units is two weeks... oh a month. So it's nice to see that now we have some data that says nine days. 00;17;44;26 - 00;17;50;03 Ken Milne That's what the data fits best. And so we're going to use nine days as the outcome. So that's a great answer. 00;17;51;01 - 00;17;58;02 Tyler James Right. And on your comment on being able to count to five, on one hand, if you learn American Sign Language, you can count up to 999 on one hand. 00;17;58;18 - 00;17;58;29 Ken Milne Oh. 00;17;59;07 - 00;18;00;04 Tyler James Systematically. 00;18;02;05 - 00;18;32;22 Corey Heitz Ken, you and you and Greg Moran must must subscribe to the same theory or the same blog posts in how to describe things. I literally, two days ago, listened to an antibiotic stewardship podcast on air cast in Greg Moran was talking about 'Why do we prescribe antibiotics for seven days or ten days?' Well, we have two hands of ten fingers and because Emperor Constantine decided we had seven days. Why do we think antibiotics work over the course of what's magically the same thing is one week on our calendar? 00;18;33;27 - 00;18;50;27 Ken Milne So. So I'm putting together a talk right now on my favorite numbers, and it's going to have five, seven, ten. Of course it's going to have 11, because this one goes to 11. But now I'm going to have to have nine on there because that's the, that's the metric we need to use for return visits. 00;18;50;27 - 00;19;00;14 Tyler James Absolutely, and again - that was a 2014 paper, but I haven't seen any other literature that uses a similar type of analysis to define a empirically defined cut point. So I think it's the best that we have right now. 00;19;01;17 - 00;19;21;28 Corey Heitz And I think it's great. I think like you said, you use data to get there as opposed to just picking a number because it happens to be how long we define a week or a month or whatever. OK, so question number three, patient level data. I understand it's important to look at patient level and encounter level data because a small subset of patients may have numerous encounters. 00;19;22;17 - 00;19;30;00 Corey Heitz However, how is this used for revisits? As that to me seems to be inherently a patient level outcome. 00;19;31;06 - 00;19;51;05 Tyler James Right. So to orient our listeners, just in case they haven't read the paper yet - you should definitely go check out the paper. But for all of the other models on ED length of stay, for triage pain scale, and for ESI score for condition acuity, we do multi level modeling, so those encounters are nested within patients but you're right, Corey, for the revisit outcome - 00;19;51;05 - 00;20;14;15 Tyler James we did not do that, which we probably should have. So again, let's go back to the context. This is an exploratory aim for my dissertation. I wasn't really sure what I wanted to do with this variable at first, and for a while we didn't know exactly if we wanted to go with 72 hours, seven days, 30 days for this cut point. Until I met with an ED researcher and she was like, 'Oh, you should look at Kirstin Rising's work because nine days is the optimal cut point.' 00;20;14;16 - 00;20;36;18 Tyler James I was like, 'Great, thank you for that.' So as I was doing my analysis and I intended to initially use a multilevel modeling approach here so that we could look at the ED revisit [as] encounter level outcome and associate that with the index revisit [misspoke: visit] I realized there was a data linkage issue. We did not know which revisit was associated with which index ED visit. 00;20;36;18 - 00;20;55;05 Tyler James And unfortunately, because at the institution I was at, I didn't have access to the medical records. So all of my data was received from an honest broker who then coded the data for me. And that had a charge associated with it - that was paid for by my AHRQ grant. And so due to resource limitations, I wasn't able to go back and ask them to recode that variable for me - that way 00;20;55;05 - 00;21;16;25 Tyler James I had those data linked. I think ultimately if we were to redo the study again or if there was replication and extension, we absolutely should be looking at revisit as an encounter level variable, umm, and trying to get more information about both the ED revisit encounter characteristics in addition to the index revisit [misspoke: visit] encounter characteristics. So I think it was the best that we could do with this data. 00;21;17;09 - 00;21;23;17 Tyler James And so we made a descriptive variable which isn't as telling, but hopefully it can encourage some additional research in this area. 00;21;24;24 - 00;21;39;09 Ken Milne Yeah, it's just one of the limitations that you have when you have funding and research costs money and resources. And you know, like you said, you went to this honest broker to do the coding and then you'd have to go revisit it. So we understand. 00;21;39;15 - 00;21;43;04 Corey Heitz This is how we had the data given to us. Sometimes it happens. 00;21;44;04 - 00;22;00;27 Tyler James And, now, it absolutely blew my mind whenever I came to the University of Michigan and what I had to request from an honest broker because I didn't have access to the data. I can go into a data portal at the University of Michigan and download deidentified patient data across all of their care outcomes for quality improvement studies, with IRB approval. 00;22;00;27 - 00;22;08;23 Tyler James I have access to that already, and that's just -- uhh, the I think the levels of research infrastructure differed a lot more than what I recognized during my PhD program. 00;22;09;26 - 00;22;13;09 Ken Milne Tyler, because there's honest brokers. Does that mean they're dishonest brokers? 00;22;14;23 - 00;22;15;18 Tyler James [laughter] I would hope not. 00;22;17;09 - 00;22;45;29 Ken Milne All right. Let's get to the fourth nerdy point. This is about length of stay. The only metric measured that was statistically different was the ED length of stay. It was 30 minutes longer in those who are deaf or hard of hearing and using American Sign Language - or approximately 9% longer compared to the non deaf and hard of hearing English speaking patients. Do you think that 30 minutes is clinically significant to these individuals? 00;22;46;14 - 00;22;55;07 Ken Milne And just a reminder, we should probably be cautious about not overinterpreting a single center, retrospective, observational data. 00;22;56;24 - 00;23;22;06 Tyler James So I'll answer the second question first on being cautious, and that is absolutely. So this work, in addition to other work that has been conducted on length of stay among deaf patients and Spanish speaking patients who have used interpreter services coming out of the University of Rochester from Jason Rotoli and colleagues. Umm, it really highlights the importance of not focusing on single site or single center and really highlights also the importance of collecting patient language information. 00;23;22;06 - 00;23;46;09 Tyler James And so let me take you back to early in my dissertation phase to talk about why we should be cautious about single site, but also the limitations of those single sites. For listeners in the United States, they might be very aware of the Patient Centered Outcomes Research network (PCORnet), which was established by PCORI. And PCORnet essentially links all patient information from unique patients across health systems that are members of the local PCORnet. 00;23;46;09 - 00;24;06;13 Tyler James So in Florida, this is the OneFlorida Data Trust or OneFlorida Clinical Network. And this links information from the University of Florida from University of Miami, from private hospital systems throughout Florida, from private clinics throughout Florida, so that we can track a single patient across their entire health care utilization within the state for those health care systems that are part of the network. 00;24;07;03 - 00;24;27;03 Tyler James Unfortunately, PCORnet, and therefore, a lot of hospital systems do not collect language status of deaf people who use American Sign Language, and so they are misclassified as spoken language English speakers. And so this creates a big barrier for this type of work. Where we want to use multiple sites, but we don't have the language variable of interest. 00;24;27;23 - 00;24;45;02 Tyler James And so that's one of the reasons we ended up using a single site for this is that we had to work with the data that we had at a site that had those data. But I also think that this really drives home the point of working with a priority population and questioning both the data generation structures of the EHR data that we so frequently work with. 00;24;45;18 - 00;25;06;09 Tyler James How are those data being collected and what decisions are being made - way before end user, such as a clinical care staff member or a receptionist get to the data? What -- how are restructuring the data from the get go, from a programing side of things? And I think that this lack of standardization does a big disservice to not just deaf and hard of hearing patients, but patients who use spoken language of lesser diffusions. 00;25;06;27 - 00;25;34;03 Tyler James I'm thinking, for instance, like Ethiopia has tens, maybe hundreds of languages that are spoken in regional dialects that aren't categorized by Patient Centered Outcomes Research Network. And so we're just not collecting those data on a particularly possible priority population and so we're gonna have to figure that out first before we move into larger site studies to really understand the breadth of this issue and the outcomes both in the ED and outside of the ED for this patient population. 00;25;35;01 - 00;25;55;22 Tyler James The first point is, is it clinically significant? And I'll start with saying, well, I'm not a clinician, so who am I really to say, what is and isn't clinically significant? But I will tell you, based off of my experience as a patient advocate and working in mental health facilities and in EDs with deaf patients, I think that it depends. It particularly depends on how we're conceptualizing emergency department length of stay. 00;25;56;09 - 00;26;16;16 Tyler James If it is being used as a patient specific metric related to care being delivered, or is it being used as a broader system level metric for crowding? If it's the latter, then it probably isn't clinically significant because if we're conceptualizing it related to the health system, is it related just barriers to inefficiency? Is it related to crowding within the ED? Crowding within onboarding? 00;26;16;25 - 00;26;48;01 Tyler James But I think that if we're using it for an encounter, specific patient level variable, it might have clinical significance, particularly for this population depending on what the communication context looks like. So if you tell me that a deaf patient who uses sign language and a non deaf patient who uses English appear to the ED for the same condition. If these are essentially matched patients, and they have everything the same except for their language. Is the communication context for that deaf patient being provided in the same way that it is for the hearing patient or non deaf patient? 00;26;48;14 - 00;27;12;01 Tyler James And if you tell me that we had major interpreter delays or we're using web based interpreting services and we're having difficulties connecting that to the internet. So harming the patient experience, harming patient trust and patient provider rapport from being sufficiently developed in the ED, I think that is clinically significant. At least to how providers and clinical staff members work with the deaf patient and how the deaf patient works with them. 00;27;12;16 - 00;27;35;13 Tyler James Is it going to have a horrendous outcome? Probably not. But I also think that it doesn't necessarily have to be an indicator of something poor happening. So, for instance, Rotoli and colleagues out of Rochester found that deaf ASL users and Spanish speaking patients who were non deaf who used interpreters had longer length of stay. But they believe that this is because it was higher quality communication. 00;27;35;13 - 00;27;52;13 Tyler James And so if we're providing higher quality to deaf patients, taking time to describe things -- interpreting processes also take longer. If that's happening, then I would say that that is clinically significant in a positive direction. And so I don't think that we can just look at the length of stay outcome and say, oh, this is higher, so therefore it is worse. 00;27;52;29 - 00;27;56;08 Tyler James We have to consider the communication context that deaf patients are embedded in. 00;27;57;14 - 00;28;37;29 Ken Milne Yeah, I think that's an important way to look at it because if they ended up having better health care outcomes because there was more, not more, but higher fidelity in the communication and maybe some of the emergency department patients. -- Like, if the length of stay was just delays and delays, or if the length of stay was one on one time with your clinician... I think a lot of emergency department patients feel rushed that they didn't get a lot of face time with their clinician to have those nonverbal communications going on as well, even if they are fully, you know, able to hear. There's so much going on in that clinical encounter and spending time with 00;28;37;29 - 00;28;50;24 Ken Milne patients. Sometimes if you just slow down, look at the patient and listen carefully, you can provide better care. So yeah, I really like that, that that just because they stayed longer doesn't mean they got worse care. 00;28;51;00 - 00;29;04;16 Corey Heitz Mm hmm. Yeah. And like you said, the sum of these minutes will compound on each other. Maybe it took a few extra minutes at the beginning to get the interpreter, but that's good for the patient. Maybe it takes a few extra minutes to get the interpreter back at the end, and that takes a few extra minutes. 00;29;04;25 - 00;29;05;08 Tyler James Absolutely. 00;29;05;08 - 00;29;16;07 Corey Heitz So it's - the care is better because you've actually taken the extra time to make sure they were communicated with appropriately as opposed to just muddling through it. And here you go. 00;29;16;21 - 00;29;39;21 Tyler James Absolutely. And I think that if we consider the variables that were and were not included in this model. Within the paper, we didn't have an interpreter use variable because what we learned very early on in speaking with our ED clinicians and clinical staff was that the interpreter use checkbox within EPIC, that was used at the University of Florida, was not used with high fidelity. 00;29;39;21 - 00;29;55;20 Tyler James And so we couldn't trust that as a variable to put into our model. But imagine if we could, then that could help us further explain this differences and length of stay. Perhaps it is because an interpreter is used and it's taking longer, similar to how we're told in colleagues found in Rochester. So, do I think it's clinically significant? 00;29;55;20 - 00;30;05;02 Tyler James If we look at it from a patient perspective, yes. If we look at it from a system perspective, I think it's just another way to indicate that maybe the entire health system, not just the ED, is inefficient. 00;30;06;07 - 00;30;13;00 Corey Heitz OK, so final question is a softball. Is there anything else? Is there any other data or themes that you want to highlight from this publication? 00;30;13;26 - 00;30;40;05 Tyler James Sure. First, I want to reiterate epidemiology and statistics 101, which is that the absence of evidence is not evidence of absence. So this single site study did not identify a significant difference in condition acuity or triage pain rating. But I think it's important to not anchor onto this single site epidemiological study working with disenfranchised populations, because this really highlights the need to encourage additional research working with deaf patients and other patients with disabilities or underserved communities. 00;30;41;00 - 00;30;59;02 Tyler James Secondly, although no differences were identified in triage pain rating with this quantitative study, we did have a follow up qualitative study with deaf and hard of hearing English speaking patients in addition to the deaf and hard of hearing ASL using patients. In that qualitative study, which I still need to finish writing up and get off my desk and onto journal editor's desk, hopefully. 00;30;59;02 - 00;31;19;03 Tyler James We did find qualitative differences in how deaf and hard of hearing ASL users and English speakers reported pain communication with providers and the clinical team. Deaf and hard of hearing English speakers describe situations where they felt that they were able to adequately describe their pain accurately: the type of pain, the location of pain, and the duration of pain. 00;31;19;20 - 00;31;41;28 Tyler James Whereas our deaf patients who use American Sign Language reported that they felt that the pain communication that - they noticed, in one patient's words: that something got lost in the cognitive process with nurses whenever they were describing their pain at triage. And for example, we used an image prompt of the pain scale rating chart that we used in the health care system, which is available in the paper. 00;31;42;10 - 00;32;11;11 Tyler James And so many of our deaf ASL using patients indicated, oh, we use this chart and they were just asking us to point to the zero to ten and then point to where in our bodies. So if we compare how deaf English speaking patients and deaf ASL using patients were describing their pain, deaf English speaking patients being very specific about the location and duration and deaf ASL using patients, not being specific about the duration or the types of pain, there may be qualitative differences in the type of care being delivered for pain management. 00;32;12;00 - 00;32;35;01 Tyler James And so I think that that's also speaks to the importance of doing qualitative work in addition to quantitative work, because we can only learn so much from these quantitative variables based off of how the data are collected and generated. And so I think it's important to continue to question if there are breakdowns in care for this population and to also identify how we are providing care, particularly with disenfranchised populations, whether that be positive or negative. 00;32;35;01 - 00;32;50;25 Corey Heitz Excellent, that's really interesting. So I wonder why and maybe you might have some insight into this, why do DHH ASL users have a hard - if that's the right term, have a more - have a different way of expressing their painful complaints? 00;32;52;04 - 00;33;19;00 Tyler James Absolutely. So so we used both the question in the qualitative study, in addition to the quantitative study, was looking at triage pain scale rating. And so in the local context of where the study was conducted, we had a previous qualitative study where we were looking at ED communication experiences among deaf ASL using patients. And many of them discussed that at triage neither the web based interpreting services were provided nor were in-person interpreter services provided. 00;33;19;09 - 00;33;39;14 Tyler James And so there was just a breakdown of communication access from the get go, many of them described, oh, it's just triage. They're just making sure that like I'm not dying. So they described like not wanting to necessarily rush the triage process or weigh it down with waiting for an interpreter. But then they're also now describing that maybe at triage there are breakdowns in pain communication. 00;33;39;14 - 00;33;58;01 Tyler James So I think on one end, we need to do better with ensuring that deaf patients advocate for access to interpreter services. But we also need to make sure that EDs have access to those interpreter services as well, whether it be a web based interpreter I like to call it the iPad on Wheels, because ultimately how we see it - or having a hot body. 00;33;58;01 - 00;34;18;17 Tyler James So having a on staff interpreter, which several hospitals in Florida and in Michigan and New York have And so I think that if we have access to equal communication or effective communication access from the get go, there will be differences in how they're reporting their pain with better pain management and reporting experiences I think among the interpreter group. 00;34;19;26 - 00;34;24;04 Ken Milne So you're going to be putting together a paper, a qualitative paper on this? 00;34;25;06 - 00;34;42;15 Tyler James Yes, I am. So the analysis is done. I mean, it's been done since I finished my dissertation. I defended back in July 2021 and I realized that if you don't actually work on revising the papers and putting them into formats and then submitting that journals, they don't do it themselves. [ laughter] And so and then life happens, of course. 00;34;42;15 - 00;34;51;22 Tyler James Right. And so now we're a year and a half into my postdoctoral fellowship and I still need to get that off my desk, but that is going to be the last paper from the dissertation study and I'm really excited about that. 00;34;52;19 - 00;35;16;19 Ken Milne I hope you'll consider submitting it to Academic Emergency Medicine because it really does have a track record of identifying these types of papers. So we've done a lot of papers out of AEM on gender inequity. We've done stuff on weight bias in medicine. And so I think that it would be a nice home for that paper to come to. 00;35;17;22 - 00;35;44;02 Tyler James Absolutely. And I'm thrilled also to see Academic Emergency Medicine really taking the charge to identify these disparate outcomes among priority patient segments, because in many medical specialties it's impossible to get deaf and hard of hearing or disability centric work within health care published in the journals. So we have to go to a disability centered journal, and so we're basically preaching to the choir at that point instead of getting it to the stakeholders who really need to have the information. 00;35;47;02 - 00;35;50;27 Ken Milne So Corey, can you comment on the authors' conclusions and compare them to the SGEM's conclusions? 00;35;52;01 - 00;35;55;02 Corey Heitz Sure, Ken. We agree with the authors cautious conclusions. 00;35;55;17 - 00;36;04;26 Ken Milne Yes, they're cautious. I love it when people don't overinterpret their own literature. So I like that they're cautious. Great. How about an SGEM bottom line? 00;36;05;29 - 00;36;26;22 Corey Heitz Deaf and hard of hearing patients present to the ED with similar acuity levels, similar triage pain scores, and had no significant difference in acute ED revisits compared to non DHH English speaking patients. However, there was an association observed between the DHH ASL patients and longer ED length of stay compared to non English speaking patients. 00;36;27;20 - 00;36;30;21 Ken Milne So how are you going to resolve that case you presented at the start of the podcast? 00;36;31;19 - 00;36;46;29 Corey Heitz Fortunately, your ED has onsite interpreter, ASL, services that can be accessed quickly and efficiently. This service is preferred over an online remote interpreter system due to technical difficulties, lack of staff training, which can lead to poor patient provider communication. You continue with normal care for your patient. 00;36;48;00 - 00;36;52;03 Ken Milne So how are you going to take this publication career and clinically apply it? 00;36;52;20 - 00;37;01;08 Corey Heitz Deaf and hard of hearing patients should be triaged and treated with the same level of concern and care. Use of appropriate interpreter services is essential as with any English speaking patient. 00;37;02;01 - 00;37;05;05 Ken Milne And so what do you tell the patient and how? 00;37;07;01 - 00;37;12;10 Corey Heitz Using the ASL interpreter, I say thank you for your patience while we obtained interpreter services. How may we help you today? 00;37;16;16 - 00;37;34;23 Ken Milne All right. It's time to announce the Keener contest winner. And last week's winner was a repeat winner. Doctor Cindy Bitter. She knew that acute pancreatitis was first described in 1652 by Dutch anatomist Nicholas Tulp. Corey, what's the question this week? 00;37;37;15 - 00;37;39;23 Corey Heitz Who first developed American Sign Language? 00;37;40;26 - 00;37;58;07 Ken Milne OK this might have two answers, we'll see. If you think you know the answer to who originally developed American Sign Language and then just send me an email. It's theSGEM@gmail.com. With Keener in the subject line. The first correct answer will receive a cool, skeptical prize. 00;37;59;09 - 00;38;14;06 Corey Heitz Now it's your turn SGEMers. What do you think of this episode on DHH patients in the ED? What questions do you have for Tyler and his team? Tweet your comments using the hashtag #SGEMHOP or post your feedback on the SGEM blog. The best social media feedback will be published in Academic Emergency Medicine. 00;38;15;20 - 00;38;29;01 Ken Milne So thank you, Tyler, for coming on the show and talking about your hot off the press publication. I guess you're going to have a long career in this space until they come up with a Lieutenant Uhura or a universal translator. 00;38;29;25 - 00;38;44;01 Tyler James Yeah, definitely. And I just hope that you know that any of the work that I do has the impact on at least one patient and like the positive patient experience for one deaf patient. That way we can get rid of systematic ableism in health care and in society. But that's a lofty goal. 00;38;45;03 - 00;38;48;01 Ken Milne And you got to aim high. Thanks, Corey, for another great episode. 00;38;48;25 - 00;38;50;12 Corey Heitz You're welcome, Ken. Always great to be here. 00;38;51;16 - 00;38;53;01 Tyler James Thank you both for inviting me. 00;38;54;21 - 00;39;19;18 Ken Milne And we have a little twist on the finale of this show. We ask all our guests to give us the SGEM tag line. But in advance I asked Tyler if he would give the SGEM tagline using American Sign Language. So I will post that to the blog post and I will share that on social media. But until then, can you give the SGEM tagline in English? 00;39;20;16 - 00;39;26;15 Tyler James Yes. Remember to be skeptical of anything you learn, even if you heard it on the Skeptics' Guide to Emergency Medicine. 00;39;26;27 - 00;39;28;00 Ken Milne Talk to everyone next time. 00;39;29;09 - 00;39;31;02 Music Track [music] Tommy, can you hear me? [music]