Subject to Interpretation

Andrea Henry on Interpreter Mental Fatigue [EP 47]

February 12, 2021 DE LA MORA Institute Season 2 Episode 47
Subject to Interpretation
Andrea Henry on Interpreter Mental Fatigue [EP 47]
Show Notes Transcript

This week we speak with Andrea Henry on her continuing work to understand and measure interpreter mental fatigue. 

Andrea has been active in healthcare interpreting for 24 years, beginning her career in 1994 as a freelance interpreter for Pacific Interpreters and later as a fulltime remote interpreter at their company headquarters in Oregon. She earned her B.A. from the University of Oregon in Spanish and International Studies with a minor in Ethnic Studies. After relocating to Atlanta, Andrea worked for six years in development of interpreter services departments for two hospital systems. She has received over 150 hours of interpreting education, attended 30+ conferences both nationally and internationally, and regularly presents on a variety of healthcare interpreting topics. One of the most satisfying aspects of Andrea’s career has been serving the families and staff as a full-time interpreter at Children’s Healthcare of Atlanta while carrying out this research on measuring interpreter mental fatigue. 

Speaker 1:

Welcome

Speaker 2:

To subject to interpretation of podcast, which takes us deep into the topics that matter to professional interpreters. I'm your host, Maria . Welcome today. We're going to find out why some interpreting encounters exhaust us more than others. Is it length, subject setting or a combination of these factors? Andrea Henry, a medical interpreter and researcher is with us today to discuss her group's research on complicated encounters and interpreter fatigue. Welcome Andrea.

Speaker 3:

Welcome . Uh , thank you, Maria. I really appreciate this opportunity.

Speaker 2:

Now you have an extensive background in healthcare interpreting, including working as a full-time remote interpreter as an in person medical interpreter. You've also held several positions as an interpreter coordinator, interpreter training, and you've presented extensively on measuring, interpreting mental fatigue. That's quite a resume. I'd say ,

Speaker 3:

Uh , it's been a great career. I , I really love it. I've had an opportunity to , uh , like you mentioned, I've done a little bit of everything and, but my, my biggest joy really is working at children's healthcare of Atlanta and carrying out this research.

Speaker 2:

Now you began working as an interpreter in Oregon in 1994 mm-hmm <affirmative> and have since then, like I said done all of these types of interpreting. Why did you want to try your hand at research? Were you bored?

Speaker 3:

No. A board . No. Um, well , uh , the , uh, the idea came to me when I was home and bored. So actually that's a good point. Um, I had , uh, I had had surgery, so I was home on medical leave in 2007 and like most interpreters. There's really not that much time while you're interpreting to look at your emails or , uh , do a whole lot other than just focus, but I was home and I was bored and I was looking at my emails and I , I was one of the , um, I was on the, the N C I H C list serves. I was , uh , a member of the NCHC , the national council of interpreting in healthcare . And I finally had time to read the emails, which was great. And this was way before , uh , Facebook. So there was really no opportunity to interact with anything else outside of that , uh , on my computer. And so I saw an email from, from Natalie Kelly, who had asked on behalf of a colleague, what, what managers were doing, what , what coordinators administrators were doing to measure their interpreter's workload. And that question , uh, really made me realize that I had an answer, but I had never thought of it. It was mostly that I'd had a couple seeds planted along the way in my manager role, where people would ask , uh , leaders would ask Andrea, what , what is the productivity of your team? You know, show us the numbers. And the numbers never really looked like what was actually happening, no one ever removed a position, no one ever said, okay, your interpreters are not productive enough, but the , that seed was planted, but it was planted. And I forgot about it. And then that question was posed by Natalie Kelly and all the answers that came in were the exact SI uh , exact same kind of calculation I had done in management, which is to divide the minutes paid. So if you look at your eight hour shift, remove your break time in the state of Georgia break, time is unpaid. So you remove that. And then you divide , uh , that by the minutes you've actually been interpreting, and then you get a , you get an ultimate , uh , percentage. And that's what every single one of these people responded on the list serve . And I remember sitting there and thinking, but that's not really the big, that's not , that's not the whole picture. A minute is not a minute. Some minutes are much more complex than others, and they require a lot more of the interpreter than others. And that's what, that's what got me started.

Speaker 2:

So did you go out and look for research or background, they might be available to address that question? Did any of it exist?

Speaker 3:

I did. I , uh, I started thinking about it because I , I responded to her , uh , you know, the email that she sent out and I sent this wildly long response on all of these different factors that could impact and interpreter's decision making and concentration and, and the , the need to sublimate disruptions. And, and I sent that. And then I, prior to actually doing a little more , uh , looking for other models or looking for other, other metrics, I, I did present at a, at a, at a conference here in Atlanta, the Sur , uh , the Southeast medical interpreter conference and Orland marque had approached me. He was , uh , my colleague at the time, but he had also been my boss twice. And then also one of my other colleagues had approached me and said, do you have anything to present on for this conference? And I said, I kind of do <laugh>. I think I have an idea. I don't, I don't know how well it'll be received. And, but I'll go ahead and, and I'll create a proposal. I did, it was accepted. I presented, and I didn't really know what to expect. I just figured I'll keep my expectations low. And I hope it'll mean something to someone. And the response was pretty enormous. There was a lot of interpreters coming up to me afterward and saying, oh my God, I never realized that this, this is why some encounters really do exhaust me. And then I had Otto, Zelman come up to me who, I didn't know at the time he's a German Spanish interpreter. And he said, Andrea, have you ever heard of Robin Dean? So he came up to me afterward and , and I said, no, he said, have you ever heard of the demand control schema? And I said, no. And he said, well, what she argues, she's a , a sign language, interpreter, researcher, and trainer. But what she argues in terms of training sign language interpreters is along the same lines of what you're saying here. Except of course I was wanting to create a measurement tool. Uh , and so he gave me the card, which I usually tell everybody, I still have it it's his card. And on the back, it says DC schema, Robin Dean . And so that was the beginning. So that was one precedent that I was able to research and look up. And I did start contacting Robin Dean to tell her, you know, we're looking at this as a measurement tool. And what you've got is a bit of a precedent and she's been extraordinarily resourceful and helpful. And then I started looking at the hospital. So I, you know, walk around and thinking, I know they're measuring things, they're measuring things all the time. They're looking at the complexity of the patient. They're measuring their vital signs to understand just how sick they are, which, you know, for the purposes of thinking about complexity, I used to call it acuity, which is more of a, a clinical term for looking at the complexity of patients. You know, how , um, how high acuity a patient might be . So in the emergency room, if somebody's got all of these vital signs that look pretty bad, like high heart rate, high blood pressure , uh , bad color, all those things point towards a diagnosis, or they at least point towards the resources needed to take care of that person. And so in the emergency department, I walked up to Laura Jones. Who's now our , uh , director of emergency services. She , at the time she was , uh , the manager, I think, of the Eggleston campus emergency department. I said, so what , what are you guys doing here? How do you , how do you measure, how do you, how do you look at how , you know, the sick the patients are and how many nurses you're going to need? How do you, how do you make sense of, of that relationship? And she sent me a bunch of information on triaging systems, which a lot of people know, but they don't realize it's, it's BA it's a basic metric, you know, whether or not something's code blue , uh , or something as simple as, as green, you know, a patient that doesn't need much. And then there's a , there's a scale that goes all the way up to the blue, but she sent me other information. And then I started looking at other scales or metrics that we have at , at , uh , at the hospital. And from that point, I thought, okay, there has got to be a way for us to translate this kind of measurement system so that they, if they know how many nurses they need, based on these details, why can't we have something that gives us a way to calculate and measure an interpreter's work based on complexity, which of course is different for us. Our complexity is a little different than a nurses , but it's the same idea. We have resources like cerebral resources that we need to employ in certain complex encounters. And so that's where it started. And then I approached Kathy Murphy, who is my main co-investigator on this research. She's a clinical nurse specialist in the cardiac unit. And she had been doing lots of research that I was aware of and I interpreted for her regularly. So I approached her and asked for, for her help. And that's where we started

Speaker 2:

Now, a project like this usually requires massive buy-in from an administration and from colleagues alike, especially something that has to do with data research and implementation. And obviously your results are only going to be as good as your data gathering and statistical protocols. You had to put a team together, didn't you?

Speaker 3:

We did. We started with just Kathy and me, and then we had a statistician Michael DeGuzman . And then my coordinator at the time , uh , uh , Jose Sanchez was also one of the team . So we had this, this team of people, all of us with our own bit of , uh , expertise. And we created , uh, the initial tool. And I will say that the tool over the last 10 years has gone through many iterations or many different versions as we've gotten smarter and realized certain aspects of the tool needed to be edited or cleaned up. But that was our initial team, but I'm a full-time interpreter. And I was at the time as well. So there's really not a whole lot of downtime for working on something like research. So most of my work was done on the downtime I created on either on the weekends or on my lunch break , um, until we were able to get some funding. So the first bit of funding that we got was from a vendor, it was from Pacific interpreters and they offered to , uh , give us a grant without any strings attached , um, without having any input on the content or, or any, any benefit from it necessarily except to benefit the field from an academic and research standpoint. And that money allowed us to , uh, have the , have the wherewithal to have interpreters involved in this, but get coverage for them. So if ever Martinez FORO , uh , assisted, which she did, she, she did the initial translation of some scripts that we created for videos. That was our, one of our first attempts to see if the tool actually performed well. So to be able to pay her to do that, that's where that grant money was used to pay , um, for interpreters to be involved in focus groups and for them to be , uh , to have the time to take an anonymous survey. So that's how the money was used. And then we also had , uh , different sets of interpreters to pairs doing what they call inter reliability, which is a , a statistical term to look at the, the way a tool performs or, and a tool is just a fancy word for something that measures something and gives you a result. So ours looks like a survey and , uh , the , that money allowed us to have two interpreters, use it , um, with the same encounter, the same interpretation assignment, and then use it independent of one another. And so then that money allowed us to let them do that while we paid to have other interpreters come and actually interpret for the family . So, so the money that we got allowed us to do a lot of this stuff, because we could pull people from the floor, which is just an expression that we use in the hospital. Like if you're on the floor, that means you're interpreting if you're in the office, obviously you're in downtime. Um , so it allowed us to do that.

Speaker 2:

So this was, it sounds like it was a development stage so that you could figure out what you needed to include in your tool. Now, once you determined what you needed to include in that tool, then what did it look like and how did you implement the first phase?

Speaker 3:

So we had a couple hiccups. We had , uh , our first statistician, Michael DeGuzman. He had left children's, he moved to New York city. Um, he transitioned to , uh , another job. And so we lost him. So we were kind of in a bit of a holding pattern until we could find a new statistician and like most things, when you look back and, and connect the dots, it was the best thing to ever happen because we found Courtney McCraken . And she's a statistician who is part of the , uh, the Emory core of pediatric statisticians and also , uh , with children's. And then now she's in a , another facility as well , uh , another , uh , organization. And she, she helped in that , that final development phase. So all of the things that I just mentioned were, were part of the development, but what she did was she looked at like, why we couldn't get over 70%, you know, 70% agreement between two interpreters using the same tool or the same set of questions is not all that sexy. <laugh> 70% sounds like what it is like a C in your math class, and nobody wants a C. And so we were really striving for over 90% and Courtney , uh , being the expert that she is with statistics and with tools was able to see where some of our problems were. And we redesigned some of the questions and just had them a little bit more concrete , uh , some of the same content, just a different way of asking a question so that people could easily answer it and have it yield the right , uh , result . So once we've finished with that phase, and we had two interpreters, Anna Maria Baracaldo and Eva Martinez fornos , they did the inter integrator reliability. And they ended up with the newly designed tool with a 96% agreement, which is really good. That means that the tool, regardless of who uses it with the same disruptions occurring during the encounter will , uh , see the same thing, 90 per 96% of the time. So after that, we were able to recruit seven volunteer interpreters from children's to , uh , use the tool, or essentially it's , uh , you can use it electronically or on paper. Most of the, the seven interpreters used it electronically. One person used it on paper, and it's essentially something you can have on your cell phone or a tablet or a computer screen where it's just, just like a survey. Like when Emory, if , if you go to Emory , uh, or Kaiser, any, any medical facility, they always send you surveys afterward, like a satisfaction survey. And you know, how you would rate the care that you got, how you would rate the survey . Uh , it looks like that it's just a series of, yes, no questions for interpreters to answer. So it's , it's very , uh , concrete and objective , uh , for most of it. And then there is some degree of Likert scales, which is , uh , just a , the term to describe whether or not you agree strongly with something or, or neutral, or , uh , not at all. It's that kind of a question. And so you answer it on your phone and then you submit and you're done. And so seven interpreters did that for their encounters, all encounters good and bad. The ones that were really the ones that they thought, oh my God, that was really exhausting. And I can't wait to use the tool and other ones that they were thinking, wow, that was textbook. You know, some encounters are just brilliant. Everybody behaves, everybody follows the rules, nobody interrupts each other. There's no construction noise. There's no, there's no , uh , double talk. There's no, no disruptions. And so we , we asked them , just use it for everything. And we got 444 encounters from that , uh , phase, which was three months. And we analyzed the data or that the statistician did. Um, she analyzed the data and Kathy and I, and , uh , Courtney, the statistician got to look at it afterward. And the most amazing thing about it was that most of our hypotheses , uh, were confirmed. And then we had also some surprises. And so that, that part, again, it just sort of it corroborated or confirmed that the tool performed quite well. It works really well. And then it also had the potential to give us information that we wouldn't have necessarily known was gonna, was gonna come out of it.

Speaker 2:

So let's start by sharing with the audience. What kind of information, or what kind of questions were asked, or which factors you were looking for?

Speaker 3:

So the, so the tool , uh , or, and it , and it does have a , an actual name, it's the CFI tool. So C F I E, which stands for , uh , complicated , uh , or complicate, or excuse me, complexity and fatigue and interpreter interpreted encounters , uh , the CFI tool. Uh , so in addition to the seven, I should mention that another 17 also used the tool , uh , through may of last year. And we just received the analysis for that. So a total of what is that? 17 and 7 23, 24 interpreters have used it at children's. And we're in another phase right now with outside hospitals. But what the tool measures , uh , are, are the disruptions that an interpreter has to attend to while they are concentrating on the message. So if you think about interpreting, it's fairly complicated from a cerebral standpoint, you're , you've got input processing and then output. And all of that involves two different languages, two different sets of , uh , cultural components that de do inform upon the interpreter's , uh , decision making . And so the tool is really measuring whether or not certain disruptions happened and disruptions are what makes it com complex things that you have to actually pay attention to whether or not you made a decision to, for example, if two people are talking at the same time, or you have someone interrupt your process, whether or not it's your receptive process or your expressive process, if somebody interrupts your process, you have many decisions you could make, you could ask them. Uh , gimme just a moment. The , this is the interpreter speaking. Can you say that again? Yeah . Just missed what, what you said, because she was talking at the same time. Some interpreters would choose in that moment to, to try to , to ignore that second speaker and continue focusing on the first speaker and, and roll with it. There , there are a multitude of ways you can handle it. And the tool does not measure that the tool is just measuring whether or not you had to attend to a disruption. And the disruptions on the tool are in three sections. So the first section is looking at communicative disruptions. The second , uh , section is looking at sensory and the third is intrapersonal and physical. And so the, and then of course there are elements as well on there , like the start time , end time looking at duration, looking at the department that somebody was in, if there's any kind of pattern around certain departments , uh , or certain facilities. And so what we did is we , we, when we got the analysis back from the statistician, we could see that , uh , one facility had higher fatigue scores and they were the same . The fascinating thing was it was the same set of interpreters that were going back and forth between the FAC the facilities. For the most part, there were several PRN that were in that group. And so one facility had higher fatigue. Um, and certain departments also had higher fatigue, like the fast moving departments and the unpredictable departments.

Speaker 2:

So it sounds like some of the, some of the situations that interpreters consider to be part part of their regular work. In other words, having to deal with people talking at the same time, having to , um, ignore the construction outside, also having people, maybe not even, you know, ignore the interpreters that the , the interpreters in the room, or maybe even attempt to speak the language. The interpreter is speaking. Mm-hmm <affirmative> that these things that we, we normally encounter and we consider to be , um, to be normal for us, can actually be quantified and can actually be taken and measured so that we understand what the effect on our cognitive load is.

Speaker 3:

Yes. And that that's the, that's the entire premise of it is that, you know, if you look at the absolute basic , uh , definition of interpreting, or just a , a really simple superficial way of thinking about what we're doing that in and of itself, just hearing information, processing it, and then , uh , transmitting it into another language requires an incredible amount of concentration, a and , and , and an incredible amount of cerebral , uh , energy. If you think about it like a gas tank. And then in addition to that already very, very complicated job. You have disruptions coming at you and those disruptions are, are things that an interpreter has to attend to. In addition to the already quite full , uh , plate that she already has , uh , in trying to take this information about a bone marrow transplant or a new monoclonal antibody treatment, or, or complications around , uh , a thyroid issue. And , uh , metastasizing cancer that , that in and of itself is a very complicated process for the interpreter. But when you're having to attend to all these disruptions to your concentration, and also the additional decisions you're making, do I stop everything? Do I not? Do I , do I interrupt for this , uh , purpose? Do I, not that in and of itself is another form of cognitive load. It's a , a decision fatigue. Um, and then there's the , the , the, the sheer fact that we have to concentrate for for long periods of time, just really excessive concentration. All of these things in the, in the literature are , are proven to cause mental fatigue. We've just never quantified it for an interpreter. And there are , uh , and , and this, I think this is one of the seeds that, that had been planted prior to Natalie Kelly posing that question. I had had a couple instances where I just over the years had walked out of an interpretation and I thought, God, that was weird. Like I'm so exhausted. And it was only 20 minutes and, and another one, and that was with one particular , uh, nurse, I won't name names, lovely person, but very, very, very complicated assignment. And then I had another one that was hours . They had no one that the , that could come and cover me. So I was there for hours, and it was a very complicated conversation about bone marrow transplant and an embryo being used, essentially to make sure that there would be a match for a, for a transplant later on , uh , in this child's life with a , with a new baby, it was a very complicated conversation and it was hours . And I wasn't fatigued. And I remember walking out of that and thinking, that's really weird. Like, why am I not? Like, if, if , if it is true that the longer it is, you should be that much more , uh , fatigued. How is it that I'm not? And so those, those little seeds were planted. And so once we got to this stage , uh , with the research, I think the, the , the desire really is, is , is to quantify it and to , to be able to calculate what kind of workload really is occurring for interpreters. And most of the time, if you think about it, this tool could be used as a retrospective kind of analysis, right? You look at the last quarter as a manager, you look at the last quarter and try to understand, okay, I had 62% productivity just looking on the surface at the basic calculation of hours, paid, divided by hours worked. But if you add the complexity factor, which that the tool would allow an , uh , a manager to take a look at that based on the interpreter's , uh , tracking with the tool of their assignments, some of which, like I said, are perfect, and some have elements of complexity, and some can really knock you over the head with adding that, that gives them a way to, to look at workload and productivity in the same way that other service lines do the ICU staffs nurses very different, or the intensive care unit is staffed. Uh , it has more nurses than a clinic environment. Uh , there's, there's one nurse to each patient generally, and maybe two patients per nurse in a clinic environment, which is very low complexity or low acuity. You might have multiple patients for, for one nurse because the complexity and the workload is lower in , in terms of what is required of that nurse and the, the, the resources that the patient needs. So it gives an opportunity

Speaker 2:

That I wanted to follow up with that. Earlier, you mentioned that the first results confirmed your hypotheses. What was the hypothesis you were looking to confirm?

Speaker 3:

Well, so we are in an external validation phase right now. And since your podcast is quite well , uh, known and people listen to it, I can't give too much information cause I don't wanna bias the interpreters in Boston, Nashville and Seattle that , uh, we , we have a group in Boston that's gonna get started on , uh , training very soon in Nashville and Seattle. We're still waiting for people to sign up. So I can't give too much information because it's very likely someone could listen to this podcast and be biased. But we have shared in previous , uh, presentations that we saw, we , we assumed certain departments would actually have higher fatigue. Uh , and that's what we saw. And we also, or we hypothesized , uh , and we also hypothesized that , uh , the certain aspects of , uh, duration would be related to fatigue and I'll leave it at that. And that was, that was the case. Um, one surprise and I can share this , uh , one surprise was that, you know, in our training, we're taught to do precessions, you know, to sort of in , in my way of thinking, it , it is to kind of prepare everybody for what I'm gonna do, and to also kind of set the stage and establish some, you know , um , some rules. And that is a , that was a question on the tool just for out of curiosity, you know, did you do a pre-session or not? And of course the data is anonymous, the statistician is analyzing it. So I have no access to it, our entire management team, which includes a variety of layers. We have , uh , Alison AOR as our manager across the entire system, we have a supervisor, surely Ocasio Reeves, and we have multiple team leads, Amy Cammel , uh , Paula Gomez and, and Susan. And so no one has access to that data. So we tried to tell the interpreters , uh , and , and assure them nobody's gonna see whether or not you did a Prees. So nobody's gonna judge you. We just wanna know if you did. And we wanna see if that might play a part in people potentially following the rules, because one of the, the, the hypotheses was that when people do not take turns and do not stay in one language and , uh, you know, do not speak at a normal tone, you know, the low talkers out there that mumble, and we can barely understand them. You know, we , we hypothesized that those would probably be associated with a higher need for concentration by the interpreter, a higher level of decision making , which would ultimately contribute to fatigue. And so we wanted to know if a Prees might make a difference. And the odd thing that we saw in the data was that the interpreters that did say they did a, Prees also had more fatigue. So it was the weirdest bit of data in, in the sense that, you know, the assumption would be, if you set the tone and you establish the rules, people would follow them. And that is not necessarily the case, but there's so much more to that data that we don't know. And of course, one, one interpreter on the team had said, you know, it's possible that, and , and this is consistent with my practice as well. Sometimes I really establish the rules when I think nobody's going to follow them. So the, the, the assignment could have inherently already been complicated, like a family conference. So when you walk into a family conference, there's a , a large group of providers, a huge board table, and you're sitting there with about 12 people. And that could be what that data means that people generally do do their precessions in situations that are inherently complex. And so we shouldn't be surprised that they were fatigued, but it's hard to know, but that was a surprise.

Speaker 2:

Well ,

Speaker 3:

It teased out .

Speaker 2:

It sounds to me also, like not all precessions are created equal.

Speaker 3:

Yes. Well, we did define it for the purposes of the research to keep it as consistent as possible. We asked them to , uh , look at a list. It was maybe maybe five factors. Uh , I don't have it in front of me. And if they hit three of those three out of the five, then they can consider it a Prees cuz we did recognize we all are sort of unique and we customize them. So that's a very good point.

Speaker 2:

Would it be fair to say that your hypothesis or your hypotheses were looking to validate a date, the kind of personal experiences that interpreters have, where they leave the setting and they go, oh my God, yes, that was exhausting. And or maybe that part of that encounter was just very, very difficult and they may not correlate it directly with fatigue, but they know that that part of the encounter was problematic. Um, so it's all anecdotal evidence that we have that we share with our colleagues all the time. Mm-hmm <affirmative> but here now we're actually trying to see whether in fact these perceptions that we have mm-hmm <affirmative> are actually producing these potentially negative results.

Speaker 3:

Yes. That's a very, very articulate way of putting it. I think, I think what we would really like to, and then , then this really does segue perfectly into what we we'd like, the, the tool to be for interpreters. So there's, there's a lot of talk about what it could mean for an administrator, which ultimately affects us tremendously, but in thinking about it for interpreters, for, for interpreters who have those stories, but don't really know, you know, necessarily what they really mean to them in their, in their mental gas tank. Um, but they share them because they need to get it off their chest. And, and, you know, you walk out of an encounter and you think, oh my God, that nurse didn't finish one sentence. It was like a half sentence, the entire 30 minutes. And it's very hard to work with as an interpreter that is not , uh , normal speech. Uh , but yet it's reality and, and it's, it's important for us to manage it the best we can. But the, the goal I think for interpreters is to have this tool, give them, give them something that really does validate their experience enough to be able to, to have it be more logical and, and not be just an anecdote, but to say, well, no, you know, of , of course it figures, you know, or no surprise there that , uh , this encounter really did require me to take a break afterward. Like to be able to move on, I think is, is , uh , is the biggest question for most of us. Am I gonna do a , a good job on this next assignment? If I don't give myself some, some breathing room. And I think, you know, that that validation, I think can increase interpreter satisfaction. It can increase our engagement in our work and that we can predict like, Ugh , I'm gonna be working with that. You know, that lawyer, I don't work in legal and I really admire legal interpreters. Uh , but you know, like if you know, you're gonna work with someone that you've worked before, you can, you can predict what is gonna require of you. So it, it can increase your, your knowledge about your needs when you're interpreting enough to, to be prepared for it. It can increase your satisfaction because you understand this particular nurse or this particular parent for me, I work with parents, mostly some of the children's facility, you know , that they might require a little more of me. And so it it'll give me a little more opportunity to be better able to predict what it's gonna require. I , I can't prevent the fact that the, the intense concentration, the exceptional decision making in certain encounters is, is gonna cause fatigue. It's just going to, but the goal with interpreters would be to increase that awareness and therefore increase satisfaction, and hopefully keep more of us in the field as interpreters in the trenches. Um, and then of course the other very big part of that is having administration have an actual metric to prove that their interpreters are more productive than the, the minute for minute analysis seems to indicate that they actually do have a , a much higher productivity, which would make it easier , uh , for providers , um, who need more interpreters to argue, you guys need to hire more. And then the administrator has the metric to prove, you know, our interpreters are working at a , at an exceptional productivity rate. This is not accept , uh , accept acceptable. It's, it's too high. They're gonna make performance errors. We don't want that. This is healthcare . And so there , there is that piece that, that validates a manager or an administrator's need to approve another position, which of course just makes interpreters that much more satisfied. If we, we have more people to share the, the workload.

Speaker 2:

Now you mentioned that you're not a legal interpreter. However, in , in legal interpreting, we always strive as a best practice to have what is known as , um , team interpreting. Uh , this is based on a few research papers, but really there hasn't been, I don't think a lot of very quantitative data that, you know, that goes as far as, as showing these results the way that your research is attempting to do so now, do you envision that this data could be used in the medical interpreting field to actually , um, send in replacements that you would anticipate that a particular type of situation and that an interpreter has been there for three hours or for two hours? And it is a , a complicated matter it's , um , a high stress matter as well. Um, you know, would a manager then say, you know, based on this information that we already have , um, you know, we're gonna send somebody to replace you in an hour in two hours, or you're gonna work together as a team so that you can both , um, have enough recovery time to, you know, do this marathon as opposed to the sprint.

Speaker 3:

Yeah. I, I think it would be highly effective to be used as a predictive type of tool in the sense that you could say, you know, based on past information from these types of assignments , uh , or with these parties , um, who, you know, do have these communicative patterns or this environment that has , uh , particular , uh , sensory issues that could be playing a part in the interpreter's performance or the, the intrapersonal section, which I didn't mention much of, but there is a component of it that is related to vicarious trauma. So, you know, if you're in a, if you're in a particular encounter, say in, in healthcare that does involve all kinds of intense vicarious trauma, maybe it's a , a rate or maybe it's a impending death , um , or maybe it's just really bad news, but not necessarily death, but we don't really know where this is going. And when you have that kind of information already, that does help management or our dispatchers in healthcare to predict just how long it would be reasonable for an interpreter to, to continue going. And most interpreters do not do not stop interpreting , uh, before they start making mistakes. There was research done, and I'm sure you know, this, and speaking of team interpreting and legal, I did use one of your well known papers in my original write up , uh , to corroborate this need. So team interpreting, I think is , uh , is very effective though. I just, we don't have enough research on it yet, but there was research done on conference interpreters years ago that showed that , uh , interpreters after a certain amount of time started making mistakes, but they didn't realize they were making mistakes until 30 minutes after they had already been making them. And that's the problem we're so focused. And we're so online , uh , term that Rachel Haring , uh , a researcher and trainer in our field uses you're so online. There's just no ability to police yourself because you're so in the zone. And so you don't often realize how many errors you're making. You might be aware of a couple, but I suspect like that research showed there's a lot we're not aware of. And so if you can predict that at a certain point, like you mentioned an hour , uh , you get a replacement that that would bode well for everybody. Again, it's healthcare . And just like in legal people's lives are at stake. Uh, you know, their freedom is at , at stake and , and legal, and you don't wanna have a , an interpreter who is potentially making mistakes that could negatively impact the outcome. And so there's a good argument for it. We just don't have the proof yet. And I would, I would be very, very happy if this tool could be used as well in that regard.

Speaker 2:

Now <affirmative> now all of the interpreters who have used the tools so far are full-time interpreters, is that right?

Speaker 3:

No, there's a variety at the children's. So at children's we've had 24 , uh, use the tool and probably the majority of them are full-time and at children's full-time is 32 hours. So anything over 32 is considered full-time, but we had part-timers , uh , use it and , uh , per diems or PRN. So those are people that might do two shifts a week. They sign up for the shifts that , um, that they want at the location that they want. And so we had , uh , we had a variety of people , uh , that, you know, represent different types of, of hours per week use it, but they were all Spanish. So that is the, that is the aspect of our data. That is wonderful, but it's also a limitation. We don't have other language pairs , uh , to show us how the tool works for them. Boston is gonna give us that opportunity. They have a , a very diverse team at Cambridge health Alliance and there's , uh , Portuguese, Spanish, and Haitian Creole, and maybe Mandarin that have signed up. I'd have to look at the list. And so that'll give us an opportunity to, to see how the tool works. It's it's in English , uh , of course, but different languages might experience , uh , different, different types of disruptions, or they might experience the tool in a different way. And it, so our question really is, does , will it perform for them as well? And so that, that data that we get in the next several months will , will give us a sense of how well it works with other language pairs.

Speaker 2:

Now, with your extensive background in telephone, interpreting and remote interpreting, do you see this tool being used by agencies, companies that have full-time remote interpreters?

Speaker 3:

Yes. You know, when I've presented over the years, people have asked me, can this tool be used for, for VR or, or P . And initially we had only tested the tool for face to face . So my thought was like, in any other tool development situation to , uh , to use it in another modality, you have to adapt the tool. And with the circumstances of last year with COVID and everybody going remote in some of these hospitals at children's, we stayed face to face . And I , I was , um, aware that, especially in some of the hot spots across the country, all the interpreters were going completely remote. And Cambridge health Alliance is one of them. They all went home and the , uh , the, it, people came there and set them up and they did VR and P from home for months. I think , uh , it's only been in the last few months that several have gone back on site and some are still working either at home or they have a center. So they have their own center , uh , of full-time interpreters and probably part-time in per diems that are part of that remote service. So the, the tool now is going to be tested and Boston will be one of the places that tests it as well as Nashville. All of the Nashville team at Vanderbilt are all still remote. So they're all doing P I think mostly, and maybe some V . So the tool will be tested for them as well. And so we'll get a chance to, so it has to be adapted and you have to add like a little menu. And my team lead and colleague Amy Kimmel is a bit of a red cap , uh , Wiz . She's just fabulous, but she's gonna just add a , a little menu that allows them to say this was P VRI or face to face so that the data can be separated out. And we could see , um , how it, how it performs with those different modalities.

Speaker 2:

Now, what about an adaptation for legal interpreting or other types of, well, I guess legal interpreting would probably be the next , uh, the next category, which could benefit from this.

Speaker 3:

You know, I've had people reach out and ask about it. I, I think it very much could be adapted to a legal setting. And the key would really be to get a round table of legal interpreter experts to not only look at the content of the tool, but to also have an opportunity to do just like what we did in the initial part of our development to get anonymous survey feedback, and then also to do focus groups so that we get a really good sense of the things that are consistent with our content and healthcare interpreting, which I imagine many of them will be very much the same , uh , across , uh , those two different arenas, medical and legal. And then I imagine there's probably gonna be some that are not on there currently that need to be on there. So it would be an adapted tool , uh , for legal, but I would, I would absolutely think it could be something that could be effective. It would just need to be adapted and have all that expert knowledge infused into it.

Speaker 2:

Now, when do you expect the next group of results to be available?

Speaker 3:

So I just got news from , uh , Vanessa Costa. She's the director at Cambridge health Alliance that her group of interpreters, the ones that have signed up are ready for training. So now we're planning to do the training. It's just a , uh , has to be consistent with the entire methodology that we used previously, which is one-on-one training. And we did get a , I didn't mention this, but we got a $20,000 grant from the duly Moore foundation back in late 2019. And we were supposed to complete this , uh , by the end of 2020, but COVID happened. And the pandemic put a hold on a , on a few things. And the silver lining is that now with the money that we received to complete this , uh, and I'm still interpreting full time by the way . But the money does allow for me to actually have some dedicated time to work on this, and I'll be able to train all these interpreters virtually. So I'll be able to do like you and I are doing right now , uh , in Atlanta, but in different , uh , locations, I will be here. Uh , but at the hospital doing the training for the Boston , uh , interpreters. And so once they're all trained, then they will get to use the tool for about three months, just , uh , one shift a week, and then we'll , um, move on to the next two facilities. So it's very likely we would have data by the end of the year. Uh , and then we would be done so that this, every time I thought we were done, then I realized, okay, I don't know enough about , uh , statistics and research and all these things. And I've thought we were done several times, but my understanding at this point is that once it is validated or used at outside facilities with a more generalizable group of interpreters, and if our data is consistent with what it has been previously, then it would be released. So the , the tool is meant to be free. We , we don't have an intention of, of selling it, but it is copyrighted. So it's already copyrighted , uh , Emory university in children's healthcare of Atlanta. We copyrighted it in 2018 and the tool would be released to managers and administrators, coordinators, directors, across the country with just a simple license agreement. So maybe 20, 21, no, 2022. Yeah, 20, 22. <laugh> ,

Speaker 2:

It's amazing to build , to think that we are already in 20 21, 20 20 , just went by like a flash didn't

Speaker 3:

It did . It did well, it went by real slow first, and then it went by fast

Speaker 2:

While we wait for these results. What advice do you have for interpreters who have encountered situations , which they feel are contributing to their mental fatigue, but are not sure whether it's, you know , in their mind or maybe they had, they didn't sleep well the night before mm-hmm <affirmative> , um, they're not sure whether to attribute it to the encounter or to outside factors. What would you say to them?

Speaker 3:

What I would say, and I I've been asked this , uh, I feel like when I presented to chia in 2019, this question came up , um, you know, like, what do we do in the meantime? And , and what do you , what are we gonna do with this? Like, how do we, how do we handle some of these , uh , encounters that cause us fatigue and then potentially affect our performance? And my first recommendation would be to, to be aware, that is the first thing. And if you are aware that that encounter did remove some of the, the , the gas tank, your, your cerebral energy, then if you're at a hospital. So I work at a hospital, I have a, like I mentioned a really well fleshed out leadership team and dispatchers, if you you're in a hospital, you're quite privileged and you have the advantage of directly telling maybe the dispatcher or whoever is sending you to new assignments is to let them know state what you experienced and say, listen, that one took quite a bit outta me. And even if you don't have the verbiage or the words to say exactly why say, I just, I just need a moment before I move on. And sometimes I do that. And like I said, we do have a bit of an advantage at children's and a , a very supportive management team. And so I can say that and have them understand that doesn't always mean I get much of a break because sometimes they're they're right , because sometimes there, there are patients waiting. There are kids that are really sick. Sometimes people learn trauma. Sometimes there's just no way to do it, but the goal is still to , to voice it. So I would say your first strategy is to just say, listen, that was really tough. Is there a possibility I could, I could go hydrate and void <laugh> because , you know, in healthcare they talk about voiding all the time, which is the weirdest thing, but that's, you know, going pee . So you can just say, listen, I need to , I need to hydrate. I need to go to the bathroom. Is that , is that possible? If it's not fine, but if it is, then, then they've, they've already started to understand at least because you've stated what you need, that they , they might actually understand that you need it, and you may not necessarily have the data yet. You don't have the tool yet. There's no way necessarily to, to corroborate it, but just saying it is, is the first strategy. And I think the other thing is to expose your management to it. So if you work with an agency and we have a bunch of agencies here in town , um, several of which are really well known and, and respected, you can also let that person know either the owner, if you're , that's who you directly work with, or if it's the, if they have a , uh , somebody who's managing the assignments, almost like a dispatcher, let them know as well. Even though with agency work, you may not be going back to back . You may be driving. You might have some time to clear your mind when you're driving, just to kinda like, you know , stop all the, the input and the disruptions that some people would argue that driving is quite stressful. So I don't know if that would be a , an opportunity to relax, but to at least let them know so that they can predict it for the next time. So sometimes this, this information, when you voice, it may not change the reality of today, but it will give them an opportunity to predict the next time. So if another interpreter that same interpreter goes back to that same provider, or if it's legal, that same attorney for that same kind of de uh , deposition or whatever it is, it does give them information to know, okay, this, this, this may be an encounter that might be worthy in the future. I don't know now of team interpreting something that gives them information. We have to articulate what our experiences are because otherwise people don't, they can't, they can't, they can't know, they can't read our minds. So, so I , my first strategy and main strategy would be just to talk about it. And, and at the hospital, even though Allison does not have the data yet, we're not using the tool as a team yet, I would say that because we talk a lot about it. It does start to permeate the way we think about how we manage workload. And it does also in a team environment, which of course is another unique thing. But if you do have a team environment, just , just let your colleagues know, wow, that provider , uh , interrupts constantly, or that provider just talks over at the same , you know, or that, that mom speaks , um , Spanglish. And it's gonna be really tough. You know, she'll say things like pu Lu , great . And you're thinking, well , I don't know , what did you ask her? She's like, yeah, losing we in . And so, you know, you have to prepare people for these elements of the, the assignment that are gonna be more complicated and that may or may not cause fatigue, but I think we need to just really articulate it. And I think that will go a long way to people really understanding just how hard this job is. And maybe valuing us just a bit more

Speaker 2:

Now, part of the medical interpreter's code of ethics, as well as , um , all interpreter's code of ethics is to acknowledge and recognize when there is an impediment to performance. It would seem to me that interpreters just need to, like you said, be aware and make sure that they advocate for themselves, that they feel that at any point they can't perform as they are expected to do. Of course, you know, you're in a medical situation and the first rule, the golden rule there is do no harm.

Speaker 3:

Mm-hmm <affirmative> yeah. And you've gotta speak up. So if we're gonna articulate that something was quite complex and that our gas tank might be pretty empty. Uh that's that's one part, but the other part too, is knowing when to say, pull me out of the game, you know, like the , the joint commission , uh , the , the whole speak up campaign. It's, it's not about just speaking up as a nurse or speaking up as a physician and noticing something's wrong here, that the patient's , uh , status is at stake, but it's, it's also, it it's the , the patient's ability too , to speak up and the interpreters, we , we have to speak up as well. So it , we speak up on behalf of the patient and we speak up on behalf of our ability to perform well enough to make sure that the patient isn't negatively impacted by our potential for mistakes. So that could be part of the way we articulate it . It's a really good point, Maria, cuz you can say, listen , uh, I recognize things may be very busy, but I'm really worried. I'm going to make some mistakes. If I don't get a moment, just gimme a moment. And, and then, then it's , it is up to us. I have some techniques myself, but it is up to us to figure out a way to refill that gas tank as quickly as possible. And I think that would that's that's another phase in the future to figure out what is it that we do besides time that allows for an interpreter whose job is so highly complex and requires so much concentration. What is it that we need to fill our mental gas tank besides time?

Speaker 2:

Andrea, thank you so much for sharing your research with us. We look forward to having you back here to discuss the next round of results as they become available.

Speaker 3:

Thank you so much, Maria. And thank you to de LA Mota Institute. I really appreciate this opportunity

Speaker 2:

To our viewers and listeners. We hope that this podcast has enrich your journey along this fascinating field of interpretation. Thank you for joining us here on subject to interpretation. Take care.