Good afternoon everybody and welcome you to our last session of this year's LHC conference and so today's topic is COVID-19 pandemic and the workforce safety. I'm Grant Chow, professor from Division of Occupational Medicine Johns Hopkins School of Medicine and Bloomberg School of Public Health and so our session will include three topics. The first one would be looking at access risk of filing the COVID-19 works comp plan by industry categories. So our first presenter is Dr. Dan Hong, a former surgeon and now is a corporate medical director at Eklund Fond Group. Dan is the excellent collaborating researcher with our Hopkins team and we have together published about maybe five papers in last five years. This topic is one of our published papers and our second speaker is Dr. Edward Bernanke, a professor of medicine at Johns Hopkins School of Medicine and also Dell School School of Medicine at Austin, Texas and he's our short leader in our research team and so I'll be the last one as a backup. If we have questions on analysis part, I'll be answering for that. Our second topic is standardizing the accommodation process for healthcare workers during this COVID-19 pandemic and our presenter is Dr. Lamisha Clear and now she's the chief medical officer at GE and also assistant professor at School of Medicine and she has been involved in health and safety at Hopkins and directly and indirectly involved in response to the COVID-19 pandemic at our school. And our co-presenter for this topic is Dr. Clarence Lam and he is an assistant scientist at Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health and he has been involved in charge of response team to the COVID-19 in our university. Because of his busy schedule, he will join us virtually and I hope the topic will work well. And so our last topic will be trying to find an easy, simple screening tool at entrance of workplace and to kind of confirming that worker is good to go back to work that day and so that's a kind of try. Dr. Clear again will be the first presenter and I will be the one doing the part of it. And so now I'll give this microphone to our first speaker, Dr. Hunter, Dr. Hum, sorry, for the first topic. Thanks Grant. I want to thank all of you people for persevering the last lecture and the last day of a meeting that's been going on for a while so we really appreciate your attendance here today. I thought it might be helpful to advance the slide. That's Grant's disclaimer. Talking a bit about background, some background about how we ended up with this database and why the medical director from AF group is here. So AF group, we're a workers' compensation insurance carrier and we operate as four divisions. So nationwide we have a pretty big footprint. We're very active in about 38 states and I've had the pleasure of being able to collaborate with our Johns Hopkins friends and Dr. Khalil, as Grant pointed out, we've pointed five or six papers over the over the years on different topics. And you know I thought it'd be helpful to to kind of go back and think about where we were. This paper was actually published in January of last year and so we started working on it the year before. You may recall, you know, the pandemic kind of started early 2020. That's when we started to see claims. January, February, a few claims trickled in. And now some of the things the paper showed might seem like common knowledge, but at that time they weren't very common knowledge. And you know at the time, you think back, it was a very unsettling period of time. And a story I've told a couple of times was in February of 2020. So we'd all heard about the pandemic. We were all aware of Wuhan, China. And I live in Michigan. We have some ski hills there. And so my wife and I had a ski resort and we got on a chair and it struck up a conversation with the person on the chair with us and it turns out she was a graduate student from University of Michigan. And so, you know, you chat with them. Oh, that's interesting. Where are you from? And she said, well, I'm from Wuhan, China. And I'm thinking, whoa, my wife and I sort of instinctively moved over on the chair a little bit. And she said, no, no, don't worry. I've been back home for over a year, but my parents still live in Wuhan. And so we got chatting with her about that and asked her, well, what's it been like? And she said, well, her parents have been confined to their apartment for the past six months. They're allowed to go out one day a week to go to the grocery store, which is in the same building they live in. But that's the life they've been living and they expect to be living that for a while. And I remember thinking at the time, I don't know if Americans will tolerate that kind of confinement very well. But as it turns out, you know, we all learned about sheltering in place and all those things that we did have to kind of learn about that confinement. And then you kind of think about that period of time also, you know, images on TV of bodies in bags in Europe, especially in Italy early on the pandemic, you know, it was a very unsettling time. And so as a carrier, we weren't really sure how this was all going to work out for us and what it was going to mean, you know, what was going to mean for energy workers, what was going to mean for our employees, about 1,500 employees around the country. And so as we thought about how are we going to manage the claims, we realized that it wasn't really kind of a typical sort of claim management that we needed to not depend on our usual claims handlers to decide whether these things are compensable or not. Because, you know, infectious disease, community-acquired diseases, typically have not been considered covered injuries in the workers' comp world. But clearly, COVID-19 was a much different kind of an animal. And so early on, we set up a team where we had actually the leadership from claims and from the legal department, from all the different areas that touch claims. And we reviewed every claim that came in, and we helped the claims handler make a decision about whether it was going to be a compensable claim or not. And so with that process in place, I think we felt very comfortable that we were trying to, you know, do the fair thing for people, but also sort of learning as we went along, right? Well, if someone had an exposure 10 days ago, was that a work-acquired COVID? You know, what were the timeframes? And all of this sort of evolved as time went by. But ultimately, we collected a fair number of claims. And clearly, I have some problem with this. There we go. But we thought that this would be very helpful, something that would be worth studying with our Johns Hopkins group. And so as we looked at our book of business, a couple of things became apparent. We needed to focus on states where we had a number of claims. And again, early on in the pandemic, you may recall, not every state had equal COVID experience. And so that's how we ended up with 11 Midwestern states. That's where we had enough volume at the time we did the analysis. And then we collected claims. And then the period of time we went from January, that's when we first received a claim. We knew the paper was going to be published in either the winter or the following early part of the year, 2021. So we had to cut off the analysis August 31. And you can see from the numbers here, how many non-COVID claims we had that we managed and how many COVID claims we had that we managed. And one of the things to be aware of in our data set is we do a lot of medical care facility underwriting, a lot of hospitals throughout the Midwest and down in the Southeast, as well as a fair number of extended care facilities. So you can imagine that we were one of the carriers that had a lot of experience with COVID claims. And unfortunately, a lot of experience with some very severe COVID claims. The flip side is, is we do not underwrite a lot of municipalities. So we did not have fire departments or police departments or those things that might traditionally be thought of as first responders. So that's kind of part of our analysis. And it doesn't really change. And Dr. Bonacchi is going to go through and explain the findings to you. But I think it's important to understand that's kind of the background. We don't write claims in all the different industry groups across the country, but like most insurance companies, we have certain areas that are our expertise. So at that point, I'm going to stop talking and Dr. Bonacchi is going to come up and walk you through the paper's findings. All right. Thanks, Dan. Well, first of all, thanks for persisting and listening to us. So we really had an issue here. Because as you know, in the workers comp market, the population, the underlying population is hard to determine, because you sell policies, and it's really a premium base. So you really don't under know, realize what the underlying population is. So, for example, let's see. Okay, so in the state of Washington, they actually determine the proportion of industries at highest risk of COVID. And you can see from this particular slide, that the percent of the Washington population that works in healthcare is roughly 13%. But in the state of Washington, which is an exclusive state for workers compensation, 37% of the claims in workers comp were submitted by healthcare workers. So you could see roughly a 2.4 odds ratio, more or less crude odds ratio for healthcare workers. But of course, we didn't have that luxury. Okay. Let's see, am I aiming at the right direction? I don't know how this doesn't seem to be. Yeah, I know. It's not very responsive. No. Okay. Okay. Well, I'm going to go all the way back. I don't know if any of the technical people could help us up here, but it's, we're not advancing the slides very well. But I'll continue on. So, so basically, insurance companies have a mixed pop. Each insurance company has a different mix of insureds. And there's no way to really look at the population at risk. So we had to come up with a new method of analyzing the claims to determine what the proportion was. You're having a problem too. Hmm. Is there any technical people floating around? No, it's okay. Well, no, it's, Grant, I'm pressing it. We need to advance the slides. Okay. Ooh. Hey, I think we might be in business. Okay, even better. Okay, well, getting back on track. So, state of Washington, they had a real population and you could determine the proportion of industries with the highest risk. So again, we did not have that luxury of this fixed population. So what we decided to come up with was the proportion. So we thought that if we looked at the proportion of non-COVID claims by industry, that would give us our baseline. So the proportion of ordinary musculoskeletal claims amongst, in the healthcare industry, compared to other industries would be the base. And the same with COVID claims. So if the proportion of COVID claims to other claims by industry, that would be the comparison. So then we would have it. So you can see we did our study January through August, with the peak April, May, for that first onset of COVID. And what we found, as you would expect, and verified by the state of Washington, healthcare and social assistance came out really high. So amongst the other claims was 34%. So as Dan was saying, they insure a lot of healthcare institutions. But 87% of the COVID claims came from healthcare. And again, at least accident fund doesn't insure a lot of municipalities. So it wasn't in the list. Now what we wanted to do was look at something a little more sophisticated than looking at just the population crude risk ratios. So we wanted to do a logistic regression, and keeping some of the variables that are confounders, such as female gender, which obviously is a predictor or a determinant of risk, age, younger age, and whether the individual who submitted the claim was in a presumption state or not. And so we wanted to adjust for that. And so what did come up was healthcare and social assurance, which had the highest risk ratio. Now, we, again, wanted to look at four digit codes within healthcare. And what we found using, again, taking our crude risk ratios, and then comparing it to another group within healthcare. Here we go. This is what we came up with doing the regression. So adjusting for presumption state, sex, et cetera, we came out with categories within healthcare that were at highest risk. And you could see medical and diagnostic laboratories floated to the top, residential facilities on down. And again, just going back, I think this kind of verifies our approach, because in the state of Washington, which basically had a real population, we came up with something quite similar. Now, I'll go all the way forward, and you could see that this is a little more verification that we're on the right track. We looked at the distribution of musculoskeletal regular claims in 2019 and 2020, and probably would have been better as a bar graph. But you can see that healthcare was 35% of all claims submitted to the accident fund group. And indeed, during the pandemic in 2020, the same proportion of claims were submitted. And you can go down all the other occupational categories and it was virtually the same. So it really tracked it very, very closely. So our conclusions are healthcare is associated with increased risk of developing COVID-19 infections and filing a workers' comp claim. In our population, the one that we studied, healthcare did not appear to consistently elevate the risk of infection with SARS-CoV-2 and filing a claim. So that's the other little proviso is the person had to develop COVID-19 and file a claim. The other finding was that presumption states do mean something. So it elevated the risk to roughly two times the underlying risk of all claimants. So a little shout out to our little research group. And this is a paper that we republished about a year ago. And it's all of us in this particular group. And we've been working together for about 20 years now doing this type of research using claims data, which is very rich source of information. And then analyzing for a whole bunch of different hypotheses to test them out. So I'll turn it over to Grant or Nimisha. Okay, thank you. So the next topic that we're gonna discuss is the standardization of accommodations process for healthcare workers during the COVID-19 pandemic. We're primarily focused on the early portion of the pandemic because that's when we really came up with this process. My co-presenter will be Dr. Clarence Lam. So hopefully he's a virtual presenter. So you'll be able to get some input from him as well. So just to provide some background again, to set the stage, we're gonna be talking about early in the pandemic when we didn't know much about the virus. And so obviously this had a dramatic effect on healthcare, healthcare workers, and their level of knowledge and their fear in the kind of work environment that they'd be in. So frontline staff especially were experiencing significant new challenges, including much longer working hours. They had increased work responsibilities and maybe different work responsibilities from what they were used to. Institutional shortages of PPE, that was something that I think all or most healthcare institutions also face. And that was adding to the complexity and challenge and anxiety among our frontline staff. And then inadequate diagnostic and treatment options for COVID-19. Again, just leading to a lot of anxiety and fear amongst our staff who were obviously taking care of the sickest of the sick for COVID patients. So Johns Hopkins did, I think, an exceptional job in trying to provide resources for our employees. We provided many resources, including PPE, adequately training on how to conserve PPE early on, because there were some shortages. And then also trying to address the emotional needs of the employees. So we had systems set up in place to try to help as much as we could. But despite these efforts, we knew that employees would still request work accommodations to either minimize their time with COVID-19 patients or eliminate it altogether. And so we really wanted to develop a standardized process in which we were reviewing these accommodations that were coming in to make sure that these decisions were equitable. And so in the first three months, from March to May of 2020, we had over 250 work accommodation requests that were submitted. And so we really wanted to create an interdisciplinary team that met on a very consistent basis in order to review all of the documentation that was submitted, review whatever literature and knowledge that we had about the virus at the time, so that we can make adequate decisions and use best practices for each request. And then part of the impetus for writing this up was really to provide hopefully the same level of knowledge and experience that we had gone through for other institutions, whether it be other healthcare systems. I know I had colleagues in smaller community hospitals that didn't have the resources, but and didn't necessarily know how to put together this kind of interdisciplinary team. And even corporate actually leveraged some of these same principles. I think the corporate sector used a lot of these same principles in order to look at work accommodations. So I'm gonna pass it over to Dr. Lam, who's gonna go through the actual process. Great, well, thank you. Happy to join all of you virtually. And just to set the stage as well, this was obviously a very chaotic period. When we look back between the months of March 2020 and May of 2020, there was still a lot we didn't know, particularly in the healthcare environment. We were still in droplet precautions at that time. This was before and the masking requirements at the institution to put in place, there was only physical distancing. We had a lot of symptomatic employees. Protocols were still being established or created on the fly. And our emergency departments were being flooded with actual patients too. The testing protocols were still in development. Tests were hard to come by. And so it was a very difficult time for our employees. That set the backdrop for an increase, a significant increase in the number of accommodations that came forward to occupational health. And so we had to come up with a process to review these and evaluate these accommodation requests kind of on the fly and what the protocols were and how did we go through a review process with that. And that's what this process is outlined here. We brought together a variety of different stakeholders. Ordinarily in the process, it was more straightforward where the request for an accommodation would come in, we would review it, we would work through it with our HR colleagues and reach out to legal. So we needed the additional input and then provide a determination back to the employee because of the volume that was coming in, as well as the high level of concern and anxiety amongst our workforce population, as well as a lot of the unknowns that were still present early during the pandemic. We brought together a multidisciplinary team that included these stakeholders that you see here, not just occupational health and HR, but also more formally legal, as well as we have an Office of Organizational Equity at Johns Hopkins that serves as the ADA coordinator. And they were brought into this process as well as infection control. As I mentioned, there was still a lot we didn't know about COVID itself. And we had never seen so many accommodation requests come to us in the past. For sense of scale, may have been a handful per month, but you'll see shortly that we were receiving many, many magnitude more than that. And so we also have six different hospitals and this process was put in place to also give us some more uniformity in how we were reviewing and evaluating all these requests across our health system. And so that someone that was submitting from a different institution within our health system would be reviewed in the same fashion as someone submitting it from a different hospital. So what you see here in figure one on the right is the accommodations process itself. And so when an accommodation request comes in, they were received by occupational health, who would then make sure that the employee and the provider that serves that employee that's provided certain information following the checklist that we'll go through shortly in the next few slides. The occupational health team then would convene what we call the COVID-19 Accommodations Review Committee or the CAR committee. And this committee was comprised of the stakeholders that I described earlier. We would review all the information that came in and that was provided by the employee as well, signed off by the provider, and then determine whether to accept that accommodation or to deny that accommodation. And if accepted, then we would reach out to and coordinate with our HR business partner to make sure that we can implement that accommodation for that employee. Or if it was denied, we would also communicate back to that employee and also work with that HR business partner too, because sometimes it really just required a level of education or assurance to that employee, many of whom were obviously very anxious. If you go to the next slide, what you see here are the components of the checklist that we had created. And so this was an effort to have some basic information gathering to compile as much of this information together as possible so that when the COVID Accommodations Review Committee met, we would have a full view of the conditions, the ailments, the medications that the employee was on, as well as any provider information. In traditional accommodation requests that come forward, there was oftentimes some back and forth with the provider to gather some more information. We just didn't have the luxury of time given the volume that was coming in, and we need to gather the facts up front so that we could make a quick determination, and hence why we pulled this together. So it goes through a variety of systems and covers some broad areas, like any kind of respiratory conditions that might preclude someone from being able to wear a mask. It includes chronic diseases like hypertension and diabetes. It also covers areas that include malignancies, particularly if someone is on some type of immunosuppressive medication. We do gather an entire medication list from that employee for that very reason as well. Any other infectious diseases that they may have or conditions that are chronic we would like to know as well, as well as any mental health or behavioral health conditions that are relevant to the accommodation request so that we can review those too. We had a guiding principle to looking these, looking over this information that was submitted, and it was grounded on the fact that it's not simply that an individual might have a higher risk of complications if someone were to contract COVID, but we would be more willing to accommodate if someone actually had a higher risk themselves of contracting COVID. So for example, if they cannot wear a mask for some reason or if they're immunocompromised because they're on chemotherapy or they have very brittle asthma or one of their medications would contraindicate, then we would be more willing to accommodate rather than due to a higher risk of complications. Go to the next slide. You'll also see information here that we gather about the employee's workplace and work environment. We do invite the HR business partner that's assigned to that office or department to also participate. We de-identify the information so that the non-clinical and non-medical committee members do not have access to the full submission or to the identity of the submitter, but just the basic information they need. The HR business partner is helpful because oftentimes they would know where that employee may be directly assigned or their work environment or who they may be working with. And so we would be able to use that information to determine whether or not somebody could properly use PPEs and at Hopkins we were fortunate that the institution moved towards universal masking even before CDC put out its guidance. And so that was protective for many folks. And so if they could not wear PPEs, for example, we would try to accommodate by having that suggestion that they wear a PAPR or that they could be assigned to their individual office or if they were wearing a respirator, N95 type respirator, that they could take more frequent breaks. A lot of this at the end was also just, as I mentioned, reassurance and a lot of anxiety and concern amongst our employees could be helped, help addressed via just some good education and reassurance from an infection control standpoint. If you go to the next slide, this is just some basic data from the period that we covered as part of the review from March through April 2020. You can see there that there is a significant increase in late March and early April where our combination requests peaked. And then it kind of dropped from there. When you look back and compile the information on these requests that came forward, Table 2 on the right shows the vast majority of these combination requests were actually due to some type of underlying chronic condition, whether it be diabetes or hypertension. And then about a quarter of them were the result of mental health or behavioral health concerns, oftentimes anxiety or depression, phobias or fear and panic disorders. And then there was also a category there of about 8 to 9% of combination requests that came in that were due to pregnancy. Now pregnancy itself is not considered a disability, so it's not an ADA accommodatable condition. But we as an institution decided to accommodate pregnancy by policy by letting pregnant individuals to not have to practice and provide care to known COVID patients. And then finally in the bottom right, there's a table that shows how these were educated at the end. About a third, more than a third of them were accepted as part of their accommodation submissions. And then there was a large chunk of them, about 60% that were pending at the time that this data was cut. However, in looking back, most of those pending were as a result of, they were in a category that needed to be cleared for denial. And so most of those pending were actually denied at the end. A lot of them, again, due to chronic diseases where the individual is not at higher risk of contracting COVID, but just at risk if they were to devolve COVID of potential complications. And so the final number that was accepted was about 38 probably to 40% that were accepted. The remainder were either denied or withdrawn. So with that, let me turn back over to Dr. Kalia for some closing thoughts and discussion. Great. Thank you. So as Dr. Lamb had mentioned, it was a very complex time. But in our experience, the provision of PPE and then a lot of supplemental education to the employees about how COVID is transmitted and how to properly protect yourself and your families. Really went a long ways and we had an internal EAP program as well as a mental health programs that were actually available at the different units for more timely interventions if needed. And in term, as well as support with our occupational health services and very transparent communication to help provide them reassurance and to help to reduce the anxiety that we were seeing and hearing a lot about during the pandemic. And especially during the accommodations process. These efforts have been paramount in terms of us being able to ensure a healthy, committed and safe workforce. As you can imagine, these are the folks that are again at the front lines taking care of the sickest of the sick. So we really wanted to make sure that they were also being taken care of. And anecdotally, our staff in the human resources department, legal and occupational health departments have really described this accommodations process as very fair and equitable. And employees have also said that they were satisfied with the process. And we have not had any appeals made to the committee's decision to date. So our work demonstrates that approaching this complex issue in a standardized fashion can allow for an organization to provide consistency, transparency and equity throughout the process of accommodations requests. So as I said, it's been kind of leveraged, I know at some other hospitals as well as corporations. So thank you. And I want to acknowledge our co-authors on the last paper who were instrumental in putting that data together. And then I can go ahead and start on this one, I think. Okay, great. So this is a really interesting paper that we did. It was the use of water and vinegar to identify COVID-19 cases during a workplace entrance screening protocol. And so just to again, kind of set the stage for how we got involved in this research. We at Hopkins have contracts and collaborate with multiple institutions, including corporations. And we had a corporation that we were working with primarily that does meat packing plants. So as you can imagine, early in the pandemic, they were very hard hit. And this particular corporation had plants all over the world. And so many areas of the world did not have, early in the pandemic and to date, did not have adequate testing supplies and did not have adequate diagnostic supplies in order to actually identify and isolate COVID-19 cases early on. And so we worked with a group of clinicians in Latin America, primarily in Brazil, Ecuador, and Chile, who had done their own research and that was conducted by the US Army in Korea, that really, you know, we were learning at that time that anosmia was a key early indicator of COVID-19 infection. And the US Army had used vinegar and water to discern olfactory dysfunction. and so they said you know we don't have access to testing for sure we don't have rapid testing we here have some PCR testing but you do have to go to a specialized center in order to get the PCR testing but vinegar we have access to so could the Hopkins team help us to evaluate this vinegar water and vinegar test and so then we can go to our leadership and say this is a validated test or because we as the clinicians here feel like it is but we again would love your support and help with that so time again anosmia was closely correlated with early COVID-19 infection and many employers at the time had this self-report questionnaire that asked about smell testing but if you look at the literature the self-report testing of anosmia or smell dysfunction is not very accurate and so it really underestimates the true prevalence of olfactory dysfunction which at the time for COVID-19 early infections is about 80% in the literature so the use of an active screen to actually decipher olfactory dysfunction was not being widely used and it really could potentially have served as a benefit in especially in regions that were resource poor and which testing was not as readily available so our objective was to evaluate an active empirical olfactory test using water and vinegar to identify COVID-19 cases and in during the workplace entrance screening and it was conducted again early in the pandemic so we're talking April through June of 2020 and it was among 4,120 meatpacking workers in Latin America so here the methods I think that there's no pointer on here but essentially these were the the equipment that our the testers would have for the employees that were walking into the plant so the the study locations were selected by the prevalence of COVID-19 and we also needed to ensure that the prevalence was high but there's also a PCR facility facility close by so that we could ensure that we're actually testing and then using our olfactory tests and then sending them for PCR to correlate the the effect of the test so olfactory testing was performed on all 4,120 employees in the meatpacking plants in Ecuador, Chile, and Brazil throughout that study period and then they were tested with RT-PCR afterwards and the next picture kind of shows what an employee goes through so the employee walks into the plant and you can see that the tester part of our team we provided them training on three major things one was how to perform the test and so and what they would do is the tester would give the employee instructions on how to properly take one swab and smell it and take the second swab and smell it and then so the the tester themselves are not standing close to the employee they're standing away but they're observing the employee actually perform the test and ensuring that they're performing it correctly and then the second part was we trained them on the disposition obviously if somebody has a positive test then you want to be able to triage them appropriately and the third thing was the documentation we trained them on how to accurately document the test since we were doing conducting this study so essentially the the employee would come in they would perform the test so a positive test is if you smell the water smell the vinegar and then you cannot tell the difference of which one is which right only the the tester knew which one was which and in that case they were appropriately triaged to follow up with health care providers and also recommended to go get PCR and all of them did and then a negative test would be that they can accurately discern the two in which case they also followed up with a screening survey of questions of any other symptoms and they were allowed to enter the plant and dr. Tao is going to review the numbers with you okay this is a form part of epidemiology 101 so basically this is a typical evaluation on a diagnostic test screening test so in this study we use the RT PCR test result as a gold standard code also it's not perfect but they use this as diagnostic standout for a COVID-19 so we also in this study use this as our judgment or gold standard to say it's a true positive or is the true negative or not so on the top we listed the positive or negative in this table are to show the result of PCR and on the left side we have smell a smell loss test also positive or negative that is the result we want to compare with the our gold standard so it's basically is the two by two table and we have important numbers listed in a B C and D and we have a total of 4,120 workers so basically in a and we have two hundred one three our worker they tested smell loss positive and also positive in PCR so we caught this true positive so in the B cell it also tested positive in smell loss test which is fine to the city and however their piece out test turned out to be negative so we call this positive for small loss test is false so it's a false positive and in the C cell is negative test result for smell loss and but positive in PCR so this negative is false so it's a false negative and in the last cell cell D and which is good part to both tests are negative so that is the true negative so ideally we want the numbers in cell A and cell D the larger the better and the numbers in cell B and C the smaller the better so how to judge this on the epidemiologists that developed quite a few indicators oh this is wrong direction okay so those indicators are generated by different ways on first we calculate a distribution by column so we calculate the vertical so for instance in this case we have 213 people out of 517 PCR test positive and are also positive in smell loss test which is counted for 41% which is pretty low and we have a term for this called sensitivity so in this case the sensitivity for smell loss test is not good and and then we calculate at on the other side so looking at those tests negative in PCR test and 85% 85% three of them also test negative in small loss test so this is so-called specificity and in our study we we think this 85 is relatively good and compared to lower sensitivity and this is a category by column this to ideal I highlighted in yellow is the most important indicators and we we can use to compel and across different studies and then we also calculate by role so so we actually calculate the indicator horizontally for instance you calculate that cell a over a and B and so that is the PPV in the left side under that role percentage so that is not the not very high and but the most important one for this calculation by row is the one I highlighted so the first of that is a 91 you see that very bright and which is negative predictive value which actually is the true negative is the D over C plus D which is a total negative in smell loss test so that means we have 91% 91% of the smell loss test and that's totally the 3,000 377 and 91% of them are also piece a PCR test positive so this is a pretty high negative value and on the other side of course that percentage of force negative is only 9% so that put back everything together back so these small loss tests have a low sensitivity that means it is not a good tool to identify Kobe 19 cases in order to do that you better do PCR or some other sense and however if we want to confirm there are known cases their health worker they can go back to the job so this probably the good one to confirm and that's why the we had that higher negative predictive value and high specificity so relatively better for this test so now I'll move back to give it back to talk clear to summarize it you more and the salute away and for that thank you oh hey this is just a summary of dr. tell she want to go through this okay yeah you can yeah okay this one okay so this table shows the or important indicators we can generate from that table and in addition to the ones who we already described and there's two others which is very which are very important one is the accuracy rate which is the overall probability that a patient is correctly classified basically is the ACL the true positive plus the true negative together divided by total number so that means the consistence there's a 20% of them different between these two tests so that 80% and the other ones also we said is a low sensitivity test but it's still be helpful to identify cases by looking at that the 3.18 which is a likelihood to be diagnosed by a small test if PCR positive okay so so three point means a threefold if you got that smell test lost as positive you still have a higher or at the three fold of like a bigger likelihood compared to those negative to be diagnosed as the true Kobe 19 cases sorry for that perfect no so our final recommendations also one one limitation of the study that is that we did not include temperature testing in this or or the inclusion of the self reported symptom data this is not what we were studying so it was not included in our data set but we also had some early indication and some early literature I think the TSA study had come out by that point already indicating that you know for temperature screening you're identifying one out of like 85,000 people that are actually covered positive so we had some early indications that temperature screening was not really a great way to to screen but and as dr. Tao had mentioned although the sensitivity of the olfactory test itself was low it can be used to augment the sensitivity of the current screening tools such as the symptom surveys that were being given in a parallel algorithm to increase the overall sensitivity of the screening process and so our final recommendations to the clients were that it could be used to augment the current screening tools in order to potentially increase the sensitivity overall of the screening process but the active olfactory testing did demonstrate high specificity so it could be helpful in knowing that there's absence of disease so the ones that are test negative and are allowed to enter the facility you have a pretty good and high negative predictive value so you're feeling pretty good about that using that test and then I'd already mentioned in order to increase the overall sensitivity our recommendations were to do a parallel active screening using anosmia potentially and incorporate it into the current screening processes that they had in place so yeah. So if you have any questions, please come to the microphone and we would be happy to answer. I think we have a few minutes left. I think the beauty of this test is that you know the result right away you don't need to wait even five minutes so okay. Hi, Claudia Hicks, OCMED in Connecticut. What concentration of vinegar were you using? Very kind of rudimentary. It was literally the Heinz bottle of vinegar that I showed you. They literally took a picture for us so that we knew. Very rudimentary. There was no specific. I mean we had the tester dip one in the vinegar, dip one in the water so it wasn't as exact as and again we were obviously working virtually with our collaborators in Latin America so I think it's that that was the vinegar bottle that was used so. And the last slide is our collaborators. The last slide is our collaborators. The very last slide. Back. If you keep going forward. It's a kick-in. Question over there. Another question. I have some older friends that both had COVID early on. They're in 88 or so and when I when we were on a conference call he for an organization he had he told us that that he'd had COVID and my question because he's a wine connoisseur is if he lost his taste for wine. My question really and he didn't but he lost this these sensation I think of salty and sour taste. Were you able to discriminate any with in or with did you run across any of those kinds of things and was it would your did your tests help discriminate those or is it pretty broad? Yeah no we didn't we didn't discern any of that. We literally just looked at the smell and we know now it's likely the the pathophysiology of why this the smell loss occurred in COVID is because of inflammation to the olfactory nerve and so we didn't do a follow-up. We could I guess but we did not look at any of the taste components. I've heard that different areas of the olfactory nerve will have different sensations. So interesting. I guess that's it and thank you very much for attending this session and I'll see you next year. you

COVID-19

pandemic

workforce safety

workers' compensation claims

industry categories

standardizing accommodation process

healthcare workers

screening tool

water and vinegar