Again, welcome everybody back to the current research and OEM and resident research presentations. This is the afternoon session. Most of you, I see familiar faces returning from the morning. And in the morning, we gave all our residents the research awards for this year. So if you weren't able to attend that, please congratulate them at some point or after the break. So I don't have any other opening remarks. I think we'll just say one more word about questions because we wanna have them on microphone and it's being recorded and it's also being streamed out to our virtual participants. So when the time comes for questions at the end of each presentation, raise your hand and the staff from the audio visual department, yes, who did an excellent job this morning. But just so you know who will approach you. We'll come by with a microphone and you'll be able to, you should ask your question into the microphone and that way it'll show up both for the audience here, but more importantly for the virtual audience at home or in the office and also later on in the recordings. So that said, we're going to welcome Dr. Nina Rodriguez from the Uniformed Services University who will talk to us about initial duty restriction for knee pain and permanent restriction following. So welcome Dr. Rodriguez. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. Okay. So well, good afternoon everyone. So my name is Nina Rodriguez. I am a third year occupational environmental medicine resident at the Uniformed Services University. Today I will be presenting my research project titled Duration of Initial Duty Restriction for Knee Pain and Risk of Permanent Restriction Among Active Duty U.S. Army Soldiers. I would also like to acknowledge my research advisor, Dr. Alan Nelson for his collaboration in this study. These views expressed are my own and don't reflect that of the USU or the DOD. There are no conflicts of interest and the data used is IRB exempt. So starting off with a little bit of background, the nature of military service involves unique occupational and physical demands such as jumping out of planes, running and ruck marching. These activities can predispose soldiers to acute or chronic overuse muscular skeletal injuries and may have a significant impact on military readiness to include lost workdays or training. In 2018, there were over 2.4 million annual medical encounters, which essentially can translate into 25 million limited duty days. So in the military, if a soldier is injured and they are not able to perform their job requirements to include their daily unit physical training, they must be evaluated by someone from the medical team who can then issue a temporary duty restriction or profile, which is a common term used in the army. That being said, accurate measures for evaluating soldiers that are being profiled is crucial to better plan for military operations and prioritize risk mitigation efforts. Overuse and lower extremity injuries make up the highest percentage according to the 2019 Annual Army Surveillance Report. On the right, you can see the breakdown by body region and then another study also referenced the knee as the third most common reason for discharge and disability. So essentially, why is this important? Personally, this was a topic of interest. Having served as a battalion surgeon in a airborne combat arms unit, I did see a lot of soldiers that did sustain or that would come in for chronic knee complaints and whatnot and that goes hand in hand with their job occupation. Also, there are limited evidence-based studies on the impact that profiling may have with long-term disability. The aim of the study was to describe the relationship between duration of initial profile for a nonspecific or overuse knee condition such as knee pain or runner's knee and assess the risk of permanent profile. Going on to the methods section, de-identified administrative and medical data on all active duty soldiers was obtained from the Medical Assessment and Readiness Repository. The study population included active duty US Army soldiers that served from January 2014 to June 2017. The study design used was a retrospective cohort including all individuals issued a first-time profile for an overuse or chronic knee condition. Case diagnoses included one of the five ICD-10 diagnoses listed on the table on the right. We excluded all individuals that already had a temporary or permanent profile for a lower extremity condition using the ICD-9 or the updated ICD-10 version that was released October of 2015. This following flowchart provides a better illustration of the study design. We did implement a lengthy washout period excluding those individuals who had, oops, sorry. Oh. You keep talking, we'll find it. Okay, sorry. Okay. Sorry. Who had, okay. Who had or acquired a temporary or permanent profile of the lower extremity using one of the ICD-9 co-diagnoses. The individuals that survived the washout period and were later diagnosed with one of the ICD-10 knee conditions and concurrently issued a first-time profile by the end of our initial observation period were included in this cohort, that being a total of 7,583 active duty soldiers. Individuals who developed a more serious complaint during this observation period or were issued a permanent profile were excluded. The cohort was then followed over a second observation period until they were either issued a permanent profile, separated from the military for a non-related knee condition, or until conclusion of the study. The outcome of interest is permanent profile of the knee. A permanent profile reflects a chronic duty limitation such as, an example is no running, though the severity of profile restriction may result in the individual being medically discharged from the military. To better understand whether duration of first-time profile could be associated with the outcome, we stratified this into seven categories and controlled for other variables such as socio-demographic data, BMI, pain level at initial encounter, total number of encounters in that first month, and also consecutive number of days on profile for those individuals that went off profile and then were later profiled a week or two later for the same complaint. Stata version 16.1 was used for the statistical analyses. We used chi-score tests to assess differences in distribution between variables and outcome of interest, along with two sample t-tests with equal variances for continuous variables. Cox proportional hazards regression modeling was then used to calculate the adjusted hazard ratio. A p-value less than .05 was considered significant. So now going on to the results. So table one shows the distribution of socio-demographic and military characteristics within this cohort. It was found that 710 individuals did go on to receive a permanent profile, leaving us with 6,873 subjects that did not. A statistical difference by gender was not found, but all other demographic and military characteristics was statistically significant as seen by the p-values. I also want to highlight marital status, older age, higher ranking personnel, E7 and above, and those belonging to a medical command, which is essentially a support unit, did have the highest distribution of permanent profile. This bar graph shows the crew distribution of subjects with permanent profile and duration of initial profile. So as you can see, the highest incidence was seen in those soldiers that were profiled under 10 days, and also those issued a profile over 36 days. And then this next table includes the result using the adjusted hazard ratio. So just to summarize, 9.4% or 710 soldiers went on to acquire a permanent profile for the knee with an observation time of 98,326 months, with a mean observation time per person of 13 months. Similar to the unadjusted distribution table, the highest adjusted hazard ratio was also seen in the older age group, belonging to a medical command, and having a higher BMI, which had not been showed in the initial table, but interestingly enough, those with the highest pain scores and individuals that did not have a pain score did have similar risk. Now looking at our main variable of interest, we used 16 to 20 days as our reference value. This timeframe was found to be essentially our sweet spot for profiling when analyzing the crude data. So under 10 days, and 21 to 25 days had the highest risk, but as you can see, after 30 days, you can observe a slight downturn. Wait, oh, am I, oh, sorry, okay. As part of the discussion, I just wanna highlight under or over profiling was found to be associated with an increased risk of the outcome. That being said, profiles that are either too short may not provide enough time for adequate medical management, which may result in re-injury, or when profiles are too long, this can delay their care in returning the soldier back to duty. Number of encounters could also be an important predictor that may assist commanders in risk mitigation efforts. Soldiers seen more than six times in that first month had a nine-fold risk compared to the reference group. I also had referenced patellofemoral syndrome and MD guidelines, which actually recommends 14 to 28 days to optimally manage individuals that have medium or heavy job demands. And so going on to strengths and limitations of this study. So various strengths of our study was that we had a large sample size, in addition to having access to objective clinical data to establish outcome versus, let's say, self-reported information. Also, by using a lengthy washout period, this allowed us with the ability to capture incident knee conditions, and this was the first study, to my knowledge, that accounts for length of profile and disability. And then various limitations include, there could have been misclassification of cases by incorrect coding or diagnoses. We also did not have the visibility to verify their medical record on an individual basis. And then also, let's consider there may be under-reporting of the injuries in the population, so those individuals that go on to seek medical care in the civilian sector, the profile may not have been transferred over to the EHR that the military uses. And then generalizability, this doesn't really apply to other branches or non-active duty members. So to conclude, the results of this study really showed that optimal time to profile soldiers, which also aligns, or there was some overlap with MD guidelines. You can also gather that profiling and consistent medical documentation are areas that warrant improvement to optimize medical care for our soldiers. I also think incorporating other military branches in these longitudinal assessments would be very valuable for future studies. And then these are my references, and I'm happy to take any questions. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. I have a question. Was it because within 10 days, initial profile was associated with the increased hazard ratio because of they got on initial profile and then it was so bad that they moved to permanent profile so quickly? Or what was your hypothesis for why that had such a high risk ratio? I mean, my impression is with, yeah, I think that they were just under profiling so they weren't allowing them that sufficient time to recover. We also did account for the running number of days on profile. I didn't include it in the second table just because there is not enough room and also time constraints. But yeah, so that would essentially take into account how long that initial profile. So if there was that one to two week break, we were able to kind of capture that data. So I just think they were under profiling them in those circumstances and maybe they really need that allotted time of like the 20 days, like 15 to 20 days was my impression. Thanks. That was a good question. I thought I saw another hand up. Yep. Thanks, really interesting study. I think it's important because with the military, just like other workplaces, there's often a lot of pressure from leadership to return people very soon, as soon as possible. And I've definitely seen problems with people returning too soon in the military as well. So my question is, why do you think that healthcare personnel were your highest risk? That's, I think, a curious finding that would seem somewhat counterintuitive compared to like the war fighters. Yeah, I mean, yeah, you would think. My impression and personal experience is, the healthcare personnel, they're usually in the clinic handling sick calls, so therefore they're not, they don't tend to do the unit PT and it's more on your own time. So I think just those other units where they're consistent in doing their daily PT, it doesn't matter what. So it was very interesting with comparing it to the special forces unit or force calm. I think that's, in my opinion, that's one of the main issues, is that we... So when you put, I think the category you put was command type? Yes. That's kind of a, that's kind of reflecting occupation? Yeah, just the type of unit, like I included. Yeah, surrogate, yeah, for occupation, okay. Thanks. When I was at USU, I did similar research on back pain and we found that healthcare personnel had the highest rates as well. Yeah. Well, I guess we need to work on that. I think I see, yeah, one question from the back of the room. Thanks for the talk. It's definitely an interesting topic. Did you look at those who were able to stay on a duty status with a permanent profile versus those who were medically separated for the knee condition? I know you excluded those who were separated for a different reason, but did you look at those who stayed on with a P2 versus a P3, for example? No, we did not for this study, but I mean, that's also something to consider is the severity of the permanent profile because obviously with the P2, they're still retainable versus a P3 equivalents a med board or them being medically separated. But, something to consider. I think, yeah, one last question there, since we've got time. When I was on active duty for 11 years, I only was required to take one physical fitness test. 19 years in the reserves, I had to take one every year. So, I kind of concur that probably the people in the medical command on active duty are not being as rigorously kept in shape. Yeah, I agree. That's a nice way to end it. Thank you. Thanks, Dr. Rodriguez. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. I'm an environmental medicine residency at the Uniformed Services University. And I'm happy to present my research here on risk factors associated with the experience of unwanted sexual contact and physical assault in the military. So my disclosure's here. The views presented here are my own, not reflecting views of the listed affiliates. And the study received an exempt notice through the USU-IRB for the analysis of publicly released de-identified data. So our background research showed that between 2006 and 2018, the estimated prevalence of sexual assault among service members has remained relatively consistent. However, there was a significant rise for female service members between 2016 and 2018. Also, between 2006 and 2018, there has been a steady rise in the number of service members reporting their sexual assault, from one in 14 to one in three. Additionally, there is an increased risk of physical and sexual assault among the LGB community as they have a higher lifetime risk of victimization compared to heterosexuals. And some theories for persistence of physical and sexual assault among female service members is military demographics, at-risk individuals, and military culture. So the LGB community may have a difficult decision of whether to conceal or reveal their orientation, as both have increased risks. Also, the military consists of a relatively young and vulnerable population. Victims may form maladaptive coping strategies, and without initiating adaptive coping strategies, there exists risk for increased level of disparities among victims. Finally, there is concern victims may face early attrition, namely in basic training and during their first tour, that decreases their ability to develop a career with the military. So our research question was, among the active-duty military, what is the prevalence of and what are the risk factors associated with the experience of unwanted sexual contact only, physical assault only, and both? And so our objectives, pretty similar, identify risk factors associated with those outcomes in the U.S. military, as well as the prevalence of unwanted sexual contact and physical assault, as well as both in the U.S. military. So our study is a cross-sectional study design with data collected through a questionnaire for the DOD, utilizing a complex survey design. Our population, we had 16,699 usable surveys that were collected between November 2015 and April 2016 for five branches of service. Our data comes from the 2015 publicly released health-related behavior survey. So this lists our variables that we analyzed. We had seven demographic variables and 15 health behavior variables. There were four categories within our outcome. That is, each respondent was classified only into one of those four groups. For example, those experiencing both unwanted sexual contact and physical assault are classified separately than those only experiencing unwanted sexual contact. Our statistical analysis was conducted using SAS-caliber sedan and included post-stratification weights, which were used for the calculations in order to generalize data to the entire DOD active-duty population. Prevalence estimates and 95% confidence intervals were calculated and reported. Univariate multinominal logistic regression was used for our unadjusted models, and multivariate multinominal logistic regression was used for our adjusted models. Our observed prevalence rates among all service members showed that approximately 1 in 10 experienced unwanted sexual contact, 1 in 14 experienced physical assault, and just over 5% of service members experienced both unwanted sexual contact and physical assault. Now, upon removing any missing outcome data, our final sample size was 15,122. So, observed prevalence showed 31% of females compared to only 10% of males reported unwanted sexual contact, and 14% of females compared to 4% of males reported both physical assault and unwanted sexual contact. The Army and Navy experienced similar but greater observed prevalence of both physical assault and unwanted sexual contact compared to the Marine Corps, Air Force, and Coast Guard. There was greater observed prevalence with the gay, lesbian, and bisexual service members compared to heterosexuals for unwanted sexual contact alone and those reporting both unwanted sexual contact and physical assault. There was also an increased observed prevalence for unwanted sexual contact with those that had a greater degree of attraction toward males compared to those that were only attracted to females. For all listed mental health concerns, we noted three to four times the observed prevalence compared to those not experiencing the health behavior for both physical assault and unwanted sexual contact. So, I utilized forest plots to show the adjusted models. I'll begin with the physical assault results. All factors analyzed relative to their reference categories are listed here on the y-axis. And each point represents the factors adjusted odds ratio with the corresponding error bars. And of course, this represents the confidence interval. If it crosses one, we found that it was not significant. Also, it's important to note that each health behavior was adjusted independently for demographics. So, we found those identifying as non-white Hispanic reduces the risk of experiencing physical assault. Of note, this is the only factor which showed a reduced risk for an outcome in our study. Those with an education level of high school or less were 2.49 times more likely to experience physical assault compared to those with a bachelor's degree or above. The results also suggest that those within the Army, Navy, and Marine Corps, current smokers, those ever reporting the use of drugs or had a hazardous drinking concern, those expressing sexual risky behaviors, obesity, trouble sleeping, and any mental health behavior assessed were at higher risk of experiencing physical assault. For those experiencing unwanted sexual contact, female respondents were 7.42 times more likely to experience unwanted sexual contact compared to their male respondents. Also, those with any degree of male attraction compared to those only attracted to females were also at risk, except for those solely attracted to males. This group did not have significant findings. The results also suggest respondents reporting a sexual orientation of gay, lesbian, bisexual, substance users, any sexual risky behaviors, trouble sleeping, and any mental health behavior assessed were at higher risk of experiencing unwanted sexual contact. For those experiencing both of these, unwanted sexual contact and physical assault, female respondents were found to be 6.26 times more likely to experience those compared to male respondents. Additionally, respondents within the Army and Navy reporting sexual orientation of gay, lesbian, or bisexual, having an equal or higher degree of attraction towards males, current smokers, and substance users, those with sexual risky behaviors, obesity, that's backed up one, obesity, trouble sleeping, and any mental health behavior assessed were at higher risk of experiencing both unwanted sexual contact and physical assault. There exists continued vulnerability among certain groups within this population, females, sexual minorities, those with mental health concerns, and several health disparities assessed. These analyses are generally in line with our literature reviewed. Obesity, a modifiable risk factor associated with physical assault in our study, is in agreement with several articles reviewed, showing either childhood physical abuse or intimate partner violence being associated with obesity as a risk factor. A later study that differed from our findings found a greater intimate partner violence victimization among veteran women compared to non-veteran women. However, there was no association found in this group between obesity and this specific form of physical assault. As for smoking, another modifiable risk factor, we found a significant association for both unwanted sexual contact and physical assault. One study reviewed assessed sexual minorities and found that lesbian and gay individuals who experienced physical assault were more likely to be current smokers. Our results show that this association may exist despite of sexual orientation. One article reviewed showed that among 28 female participants, sexual activity increased following the experience of sexual assault, but alcohol use also increased. We found hazardous alcohol intake to be an associated risk factor for unwanted sexual contact, but it may also serve as a maladaptive coping strategy for those who experience sexual assault, and partially explain the risk of those reporting sexual risky behaviors for the outcome of unwanted sexual contact. Although our data analysis suggests that there is some association with these risk factors, our study cannot speak to temporality. That remains unclear. So, limitations of our study include recall bias, which is inherently a factor when using a self-administered survey, and misclassification of exposure may exist in sexual minorities who choose to conceal their orientation in efforts to minimize discrimination or victimization. There was a low overall response rate of 8.6%. Furthermore, these results are not generalizable to the general population, and we analyzed lifetime risk of many characteristics and outcomes versus their in-service risk. Our strengths include a large sample size, allowing for greater precision of estimated prevalence, the inclusion of five branches of military service under one data set, we reduced the risk of collinearity by adjusting each health behavior independently for demographics, and weighted methods helped to adjust for the low response rate and allow for generalizability to the entire DOD active duty population. So, in conclusion, we found that one in 20 service members experience both unwanted sexual contact and physical assault. And although significant efforts have been made to address these outcomes in the military, several subpopulations remain at risk. Our analysis of multiple modifiable health behaviors show elevated risk outcomes assessed that can serve as indicators, which may assist providers in identifying possible trauma and initiating adaptive coping strategies. Our study analyzed a broad range of health behaviors, which provides significant areas for future focus and research. One of which that we're looking into with the new survey that's out is to investigate those experiencing unwanted sexual contact during their active duty service compared to their time prior to service. I want to thank you all for your attention, and I'd like to acknowledge the following individuals who provided significant support and guidance throughout my research. Thank you. I have a lot of references. All right. I see a couple of questions from the back. Thank you, Dr. Espinoza. Very well put together presentation. My question is, every time you see LGBTQ, how come the T wasn't in there for transsexual? Is that just from the survey data that wasn't asked? Also, throughout the study, were there any other risk factors that you wish were assessed that could have lended to a more targeted campaign to help prevent some of these assaults or unsolicited sexual contact? For example, did it happen in the barracks? Does it happen in the locker room or something like that? That is a couple of great questions. In terms of the transgender, there are questions that were asked within the health-related behavior survey, but that data was exceedingly more difficult to actually acquire with our data set and to analyze. Unfortunately, we did not have the opportunity to do that. We had taken a look at the long list of risk factors that we did assess. It didn't go as granular as you're speaking to as who is actually staying in the barracks or living off base or on base, but not on barracks, on military housing. That would be great to analyze that data as well to see if there is increased vulnerability or increased risk within those populations. We did have approximately the same number of risk factors when we initially analyzed this data. We added a few more we thought were pretty important to have, more so with suicide ideation, suicide risk. It's a comprehensive list, but there's always more that we can do. I heard a whispered comment from a colleague in the audience. Some of my military peers have also commented that sexual trauma and violence in the military is often underreported. I was wondering if there's any way that you can estimate the degree of underreporting in your own data set. If you are able to speak to whether or not that confounds your results at all in this really important study that you've done. I believe there is a significant amount of underreporting. One resource that we look at takes a look at what the actual number of reported cases are and then takes a look at a survey that's put out to show the two numbers and see whether or not they balance. Based on that actual data showing the number of reports and the data that's coming from the survey, there is a huge discordance there. There's been efforts though by the military, as mentioned, to try to minimize that, really dating back to about 2005 where they have programs set up to try to make it easier for the victim to approach somebody and speak to them. Military service members here will know of SAPR and victim advocates and such and the ability to either do a restricted or unrestricted reports which might provide some level of comfort in terms of bringing in a legal aspect to the reporting versus not. The effort continues to try to do that. There is a discordance there for sure. Hi. Very cool study. What do you plan to do with the results? Are you going to recommend that physicians ask these type of questions in general reviews or how do you want to move these results into actions? I think that is something that I'd be interested in. That is a recommendation that we had put here in this presentation. I plan to put that into our manuscript and hopefully publish at some point soon. If we have these risk factors available to providers and perhaps they can recognize those and at least think about the possibility that sexual trauma or physical trauma had occurred, you can perhaps spend another minute or so and try to find out if there is anything more that can be done to provide resources so that there are adaptive coping strategies that those victims can use. I don't think I need a microphone. You have to use it as the earphone. Thank you very much for this presentation. I was just wondering, did you collect any data on the questionnaire on who the reported perpetrators were, i.e. was this an intimate partner, was this a superior officer, etc.? The questionnaire didn't offer that data in terms of perpetrators except for one question that gives us a little bit of information on that question. That's in terms of who was actually experiencing this in terms of their attraction to a certain sex. What we found is that it tended to be those that had an equal or higher degree of attraction towards males. our literature review goes along with that, where males tend to be perpetrators in both heterosexual or LGB community relationships. It tends to be males that do. And that's really as far as the questionnaire went, but that is an incredibly interesting question that deserves more attention. I think we'll give you a breather from right there. OK. All right. I understand. Great. All right. Well, thank you very much. Yes. All right. OK. OK. OK. OK. All right. Thank you very much. Thank you for the questions in the audience, as well, here. Occasionally, we get a medical student who sends in an abstract. And fortunately, we get to hear them sometimes there. I'll say the last medical student I think we had that got an abstract on the podium was Alberto Kevin Martinez. And he's done very well. So our end of one study implies that you'll have a great future. So let's welcome to be doctor, Ian Shoneman, who will talk about cold weather related injury events in OSHA severe injury reports. Come on. So you can just kind of point it at that. And that's forward and pull back. Thank you. Awesome. Thank you very much. So my name's Ian. I'm a fourth year medical student at GW in Washington, DC, graduating in about a week and a half, for anyone who is counting. Congratulations. Thank you. Thank you. So thanks so much for the wonderful opportunity to present at AOHC. Before I begin, I would like to thank Dr. Weaver, Dr. Tustin, and the rest of the OOMN team at OSHA. Given that I'm a medical student at a place where the curriculum really doesn't have much oc med stuff in it, I can honestly say that they have taught me everything I know about occupational medicine. All right. So now, cold weather related injury events in the OSHA severe injury reports. As you guys have seen for every other presentation, I'm going to be talking about the cold weather As you guys have seen for every other presentation, we have a nice disclaimer slide. We have no disclaimers to report. All right. So now for a little bit of background and why this project actually just got off the ground in the first place. So as you may remember last fall, the Biden administration made it a priority to begin drafting new rules and regulations about workplace injuries related to climate change. As these New York Times articles discuss, these prioritizations reflect a growing recognition of the health threats, and in particular, the workplace hazards of climate change and global warming. My month long rotation with OSHA took place in September 2021, just as this hot topic was receiving wider news coverage. So it's obvious that climate change associated with rising global temperatures has a clear effect on an increasing number of hot weather related workplace injuries. But climate change's impact is less clear regarding change in cold weather related workplace injuries. Now before we get into any debate on climate change or anything, there is research to demonstrate that warm Arctic episodes, episodes that may be associated with climate change, are linked to an increased frequency of extreme winter weather in the US. So in conjunction with research on workplace injuries associated with hot weather, we began this project hoping to describe cases of workplace injuries related to cold weather. So to describe cases of workplace injuries related to cold weather, we turned to OSHA's severe injury reports database, also known as the SERS database. Since January 2015, employers have been required to report instances of severe workplace injury to OSHA through this database. The workplace injuries included in the SERS database include hospitalizations, amputations, and loss of an eye. Importantly, though, only employers located in states under the authority of federal OSHA are required to report workplace injuries via this database. In the map seen here, the states for which we have data in the SERS database are the states that are colored in the lightest blue color. For example, Pennsylvania, Wisconsin, and Texas. There we go. All right. So we can now take a closer look at the SERS database to see how we analyze the cold weather related data. The data we analyzed covered the time period between January 1st, 2015 to January 31st, 2021. The columns of the database are organized by occupational injury, illness, and classification system codes that we use to identify records that were likely related to cold weather. To be more exact, we sorted through the data based on nature, event, source, and secondary source codes. For example, the source code 9261, as I'm sure you all know off the top of your head, refers to temperature extremes environmental, cold environmental. We then manually reviewed free text narrative fields to identify and remove non-weather related injuries. As I'm sure you can imagine, there are many frostbite injuries unrelated to cold weather, but are instead due to work in freezers or handling cold objects. Since injury data is manually entered by employers, there were occasionally errors in which these frostbite injuries, which were unrelated to outdoor exposure, were still labeled as environmental. It was necessary to review these free text fields to ensure that we only included cold injuries that were indeed related to cold weather. To better analyze the SERS database, we wanted to calculate the incidence of cold weather related injuries within the overall data set, within plant hardiness zones, and within individual states. Plant hardiness zone, also known as PHZ, is a term defined by the US Department of Agriculture to denote the average annual extreme minimal temperature in a specific area. Using plant hardiness zones, we could better analyze how cold weather injury events aligned with the extreme minimum temperature in a region. Here, you can see a map that demonstrates the various plant hardiness zones throughout the continental United States. The zones are color coded, and they denote ranges of 10 degrees Fahrenheit. You'll continue to see this chart throughout the rest of the presentation as a way to visually compare the incidence rates to plant hardiness zones across states. Separately, to calculate incidence rates within individual states, we used population and employment data downloaded from the US Census Bureau. Finally, on to results. The CSRS database included 62,183 total injury events between January 1st, 2015 to January 31st, 2021, 1,330 of which, about 2.1% met inclusion criteria. As you can see from the chart on the left, the majority of these, about 98% were due to falls or slips. 13 of the 1,330 injuries included amputations, all of which were finger amputations, while 1,314 involved hospitalization. The chart on the right shows the 1,330 injury events by month. From 2015 to 2019, the average number of cold weather-related injuries was about 230 injuries per year. This average decreased to 151 injuries in 2020. Note that we only have data through the beginning of 2021, just in January 2021, so we were unable to calculate an average for the entire year. Here, with the figure on the left, you can see the cold weather-related injury incidents by state. The highest incidence was in South Dakota, with an incidence of 1.26 cold weather-related injuries per 100,000 workers per year. The overall incidence rate of cold weather-related injuries was 0.21 injuries per 100,000 workers per year. Again, note that only states in which private employers are covered by federal OSHA are included in the study. As such, all the states you see there that are grayed out represent states that do not enter information into the CSRS database and are not included in the study. It does not mean that those states don't have any cold weather-related injuries. Next, with the chart on the left, we have the cold weather-related incidents by plant hardiness zone. As you can see, the cold weather-related injury rate aligned closely with plant hardiness zone. The rate of cold weather-related injuries was highest in the coldest plant hardiness zone in which any injuries occurred, with an incidence of 1.0 injuries per 100,000 workers per year. However, these injuries were not limited to states or plant hardiness zones typically associated with cold weather. Cold weather-related injuries were reported in all plant hardiness zones occurring in the continental United States. There are a few conclusions that we drew from this project, with this slide noting one of the more surprising aspects. We found that cold weather-related injuries actually make up a similar percentage of the CSRS database at about 2.1%, as do heat-related injuries at about 2.5%. Now, this 2.5% heat-related injuries percentage actually comes from a presentation from last year's AOHC, heat-related acute kidney injury and indoor and outdoor workers in the US. And as that title suggests, these heat-related injuries actually include both environmental sources of heat and indoor sources of heat, aka heat-related injuries that may not actually be due to hot weather. So that 2.5% may actually overcount the percentage of hot weather-related injuries. With so much focus on how climate change may affect hot weather-related injuries, it's clearly important to also focus on cold weather-related injuries as well. In addition, as you can see in the top chart, there is a clear decrease in the raw number of cold weather-related injuries as we move from 2019 to 2020 and the beginning of 2021. We believe that the decrease in the labor workforce due to the COVID-19 pandemic could be a cause of the decreased cold weather-related injuries seen in 2020. Also, while the cold weather-related injury incidence increases, as the plant hardiness zone temperature decreases, these injuries are not isolated only to cold weather areas. As you can see in the bottom figure, there were cold weather-related injuries in every single state in the Deep South. It's clear that there are many variables other than the average annual extreme minimal temperature that affects the incidence of cold weather-related injuries. Some of these variables might include preparedness for cold weather events, number of anomalous cold weather events, or ice versus snow as the predominant factor in a weather event. An obvious area for future study is a continued look at the CSRS database, which updates each month. Especially as the workforce returns to in-person work, it'll be interesting to see what happens to the number and incidence of cold weather-related injuries. In addition, it would be interesting to determine how cold weather-related injuries are correlated to major climate and weather disasters. NOAA, for instance, has many databases, such as the Snowstorm Database and the Billion Dollar Climate and Weather Disasters Database that could be used to help study any connection between these injuries and specific weather disasters. Thank you again for the opportunity to present and share our project. I'm happy to answer any questions that you guys might have. Thank you. Thank you. I see one question up here first. Before I ask your question, I'll just ask you where you matched so that we can start to poach you and all the program directors. Headed to Hopkins for internal medicine in July. Underachiever. Thank you, Ian, for an excellent presentation. I was wondering if this database would allow you to analyze the data by industry or occupation. And the reason I was wondering that is whether we can see if these are, you present a lot of data about how most of these injuries are slips, strips, and falls, and whether there's any relationship with whether people were able to not commute to work from home during COVID-19. Thank you. Yeah. So I believe there are a few columns that are based on the industry. There are very broad swaths of industry. So I'm not sure how granular we could get, but that's definitely something that we could do to further study this. Another? Dr. Sal? Between 2015 and 2016, it dropped significantly. Which was different from your other trend, which was trying to show that climate was making a difference. Is there an explanation for that? I don't have an explanation for that. And I am sorry that I don't have the standard deviations for all this stuff, because I think that would probably be helpful for this. Because while there is a decrease when I calculated standard deviations, and so because it's not on here, you have to kind of take my word for this. Don't worry. We won't. Yeah. I don't think it was enough to actually register a huge difference for at least the winter months that you see there. Right behind. So looking at your data, that's definitely something we would like to use to help prevent slip, strips, and falls. I was curious if just looking through the data, there is anything you're able to look further into. And I'm not sure if it's something Just looking through the data, there is anything you're able to look further into that we could apply on a practical level now, or if that would just be something we should look into in the future. Yeah. So we started to look into that. And by started, I mean, I was looking in the free text narrative fields, which are very difficult to analyze. But they do give a lot of information about how this stuff happened. Because that's obviously, I mean, we're not just doing this for fun just to see what happens. The whole point is then to try to prevent these. But yeah, that was the free text narrative column was the best way we could try to figure out why these happen and how to try to prevent them. Hey, my question is regarding just the severity of the diseases. So what qualifies somebody getting in the SERS database versus not? And I'm interested, too, in kind of, not to put you on the spot, and I know this wasn't part of it, but what also qualifies for heat injury? So just looking at what's the severity of the cold versus the heat that make it into that database. So the SERS database, it really just includes the hospitalizations, amputations, and loss of an eye, which is important because it doesn't include any deaths or anything like that. And also for this data, it's probably important to note there were things like MIs, which did not come up frequently, did come up, and we did actually not include them in this because the thought of including an MI is more exertional based, and that would have happened shoveling snow or walking up like two flights of stairs. And so we decided not to include that in here. But in terms of the severity of these injuries, obviously it's severe enough to put you in the hospital, but there are no deaths included in this. Great. All right, thank you very much. You'll be hearing from most of the program directors in a couple of years. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. All right, thank you very much. Our last speaker is actually, I would say, he's familiar to the podium except he presented last year, and there was no podium. And so you get your chance to do it in person. Dr. Samuel Koh comes to us from the OEM residency program at the Harvard School of Public Health, and he'll talk to us about health and fitness in the cohort of firefighters in their earlier career. So welcome once more, I guess I'll say. Thank you. Come on. All right, I'll set this right here. So sort of point at that stuff over there. That stuff over there. That's forward and forward and back. OK, OK. All right. Thank you for the opportunity to share some of the work that we've been doing. We'll jump right into it. I'm almost crawling my way to graduation, so we're getting there. So obligatory statements, we have no financial interest to disclose. We were funded by the ERC, which we're very thankful for, as well as additional support from the Kennedy School. We were working with human research, so we did go through IRB approval. And so jumping right into it, thank you to Dr. Blonin for providing a very good introduction to the firefighters. And hopefully I'm probably preaching to the crowd on this at this point, but we can all agree that firefighting is very dangerous. But what we're concerned with is the prolonged sedentary periods between actual activity of firefighting or even high-risk training. And with their multitude of exposures, we can see that they have excess cancer risk, PTSD. But as pointed out earlier today, obesity is in 3 quarters, or being overweight is 3 quarters of these individuals, and obesity being at least 1 third. And depending on which study you find and cite, these vary a little bit, but these are very striking numbers. And for us, we were very interested in the fact that cardiovascular disease becomes a greater concern. So as they age over their careers, they develop hypertension. And one study finding that they're having increased risks of cardiomegaly. And so when they are finally called into action and arrived to the scene, they faced a greater risk of sudden cardiac death. And that contributes to a leading cause of deaths on duty, with those having an established coronary disease being at 15 times risk of that being the cause of their death. And so the question we ask is, where does this all start? At what point do we get there? So we take a look at their training pipeline. To even begin, you have to pass a medical exam, which has standards and physical assessment tests. Each state might have slight variation. Massachusetts has their initial higher requirements. And as you look towards their training, it's a 12 to 20 week long course that follows a standard, NFPA 1001, which, again, has prescribed requirements for physical training and skills training. So then as we follow them on from graduation into their early careers, we're particularly interested for this study, their probationary period, which can be up to one year or even two years for some of these individuals. What happens then? Following on some earlier work that was done, what we did see was they tend to practice rather sedentary lifestyles. And their dietary habits just declined from graduation, unfortunately. And then looking at it in a previous study, it did find that a Mediterranean lifestyle has been shown with a lower prevalence of hypertension and increased aerobic capacity. They also found that the recruit training itself improved fitness. And then when following a Mediterranean lifestyle, They did better in terms of fitness, hypertension as we mentioned, and part of the reason that diet was chosen is that it is of the healthy options for diets that was a preferred option among career firefighters. And so to look at kind of the general lifestyle, we used a many lifestyle score which is composite between a look at their diet, their BMI, which I know is not a fan among the crowd, but that's okay. We also looked at, you know, tobacco use history, physical activity, and other behaviors such as how they slept, what they did on their off time. And so based on this, our hypothesis was that existing evidence of unhealthy firefighting working environments, we would expect that these new firefighters' health and fitness will continue to decline from the academy graduation into the period extending beyond even their probationary two years. So our study design used a prospective cohort and since we were drawing from a previous study, we had 92 firefighters that were enrolled and we had already followed them for eight months, and now we were following them continuing past their two years. On average, we were able to catch up with them again at about 32 months. Unfortunately, we did have quite a bit of loss to follow up down to 26 out of the original. Our outcomes of interest were carried on from our past work. Looking at body compositions, we did use BMI and then using bioelectric impedance as a measure of body fat percentage. For physical fitness, we used a one-minute push-up capacity as a surrogate. Lifestyle, again, following the many lifestyle score. And then a number of mental health scores, Beck's Depression Index, PHQ-9, PCL-5, which the last being the PTSD more specific. And then looking at the resting blood pressure, systolic and diastolic. So the data we present here are only including those that were available and actually completed each of our time points. Statistical analysis, we were really just trying to compare the overall group between each of the time points using rank-sum tests for the whole different periods, and then a Friedman test between the paired groups using a Bonferroni correction. All right. So now we get to the data. What do we find? So as we saw, or in some ways expected, starting from during even their training period, let's see if we can get the Oops, that was not what I wanted to do. There was a laser pointer button. Where did it all start? Okay. So even from their start to graduation, not a huge change in their overall lifestyle, but as you can see from continuing on, we did see a significant decline in their overall lifestyle score. BMI, just starting from the beginning, you know, we start seeing kind of a return to their baseline, you know, starting here at 28, going down to 27.6, and then actually coming back up to 28.4 as we go through the three major time points, respectively. And looking at body fat percentage, we can see that the training period had an improvement. While not statistically significant, you can kind of visualize a trend, but what's striking is that, you know, after they leave that training period, we're kind of right back where we started. What was interesting to us was as we looked at their blood pressure scores, again, we couldn't quite achieve significance, but, you know, throughout that time, their blood pressure was already increasing, and then for diastolic, they're already going up and then holding steady. And then looking at pushups for our, you know, surrogate for overall fitness, or more specifically, our aerobic fitness, we do see that their training does indeed provide a substantial benefit, but that benefit is not sustained, and unfortunately, we have an attenuation of that going forward into their careers. Their PTSD depression scores did not really show any significant change. They were pretty stable throughout their time period that we were observing them, and that includes the PHQ as well. So then, as we consider here, what could we say about this? You know, unfortunately, we didn't quite have enough time points or enough participants to show the significance that we wanted to in some of these scales, but we do see a trend of attenuating beneficial effects from that training that we were hoping would give them some substantial or sustaining improvement in their health. And even in that probationary period, you know, we're going past it where they're no longer that proverbial new guy with the new guy stressors, and they've gotten acclimatized to their new profession. They aren't able to bounce back to getting back to their healthy diets or healthy lifestyle that they had picked up, presumably, during their training period. So one other thing we noticed, but were unable to really look at more in depth, was the fact that between the different fire departments, there were differences in their lifestyle and their overall culture and the way they viewed fitness. And I think that's something that we can see between the different branches and other services, as well as law enforcement. That would be something that we'd want to be able to look at a little bit more and consider further, but, you know, we would need far more participants from each of the different stations. And so one of our major difficulties was, you know, the selection bias. Everyone volunteering, we did provide, you know, a gift card for those to come back and participate with us. But unfortunately, we weren't able to get very many people. And then during this period, we also ran into COVID. And so we were unable to get as many people to participate. We were unable to travel to obtain data from many of these fire stations at the same time. But, you know, overall, I think it's pretty well agreed that firefighting is dangerous. But the one thing we sometimes may forget is that a sedentary lifestyle is a hazard of its own. And especially when they're stuck waiting around for something to happen, that's what they revert to. And with that chronic burden of disease and then being suddenly called into action, that puts them at a great risk that we should definitely continue to consider research. If we were able to, we would love to continue this research, trending them, you know, for five years or even into their late careers, to see really where that cut off and the change happens from what we see in kind of the more, you know, unhealthy cases that might come for fitness, for duty or something along those lines. And, you know, more importantly, what sort of interventions could be could be done. Unfortunately, what we couldn't present for you today was a parallel study that had looked at putting in an intervention in. But that cohort, we were unable to catch up with them, unfortunately, due to COVID. So. All right, I do want to thank Dr. Kales and Dr. Land, who were very instrumental in bringing us onto this project and continuing for it. And then, of course, all the firefighters that took their time out to participate in our study. Thank you. I'll ask you one question before we start there. I know you had some losses to follow up. Is there any reason to think the losses to follow up were any better than doing any better than the you know, usually you get sort of volunteer bias and they're the ones who are better off. Do you have any data or information on them? No. Unfortunately, we didn't we weren't able to establish that. But presumably that's that's a concern. And, you know, that would that would probably. Yeah. It would bring our results towards a null. Unfortunately, you're probably biased towards the null as a consequence. Right. Dr. Koh, big fan. Great talk. Thanks. We're on the same page. My question is, did any of your metrics include an evaluation of sleep quality? We know how important sleep quality is and it's known cardiovascular risk. And so sleep quality, not exactly, but their sleep amount duration. That was part of the MediLifeSize score. Great topic. Appreciate the information. Did you as far as your study population goes, were any of these volunteer firefighters versus full time firefighters? And then within that, it's very common for firefighters to have second jobs. Did you do any subgroup analysis as far as those who might be working in their off time versus those who don't? My understanding was that these were all full time firefighters. But whether they did participate in any off duty occupations that we did not inquire. But that's good. That's a good point. Dr. Hodgson, I'll have a question in a moment. Thanks. Great study. Questions always come up around lifestyle and then the context is culture. Volunteer professional and paid professional firefighters, you know, each station is different. Do you have any metrics about how to measure culture in the different firehouses and how that influences whether people stick to this, that or the other? I think the unfortunate short answer is no. To measure objectively a culture, I don't have anything that comes off my head. We collected from many fire stations across New England and we had many hours of driving as a team thinking about these questions. And we could not for the life of us think of a good way to study their culture or to, you know, to somehow put in some sort of objective measure other than saying, you know, if we had more individuals, let's say you took, you know, we were able to get, you know, substantially a greater number from each station and you can look at their different lifestyles. Then you can maybe say there is a difference between their station cultures. And then consider, well, does that match up with what their other outcomes are? But I think that's as far as we were able to think of. But if you, I mean, if you have an idea, we would love to take that. Dr. Salmon. John, you shouldn't have let me in for this one. You shouldn't have let me here make comments on this one, by the way. A couple of things. First of all, I think you missed the baseline. I think the baseline is really long before they get into the academy. In order to get into the academy, they have to pass a CPAT. To get into the CPAT and to get, to be able to accept it, they all get in shape for that. So, their baseline is before that. So, by the time they get to the academy, they're all in better shape than they've been for most of their lives. Then they go through the academy, which is incredibly rigorous. And so, they all lose weight and they get strong and that stuff. And then they go back to their baseline. So, I think that you missed the baseline. The other thing is, what's the point? Because Stefano Scale's literature clearly says that firefighters don't have more coronary artery disease. They have more coronary events at work because of the sudden, you know, work stress. But an old ER doc, let me tell you. First time it snowed, lots of cardiac events. All the people with underlying cardiac disease go out and shovel snow. So, I think that part of the problem is I think you missed the baseline here. That's a point well taken. I think to piggyback off of that, what we did also notice, not from doing this study per se, but from actually just doing initial higher exams in our clinic, a lot of these were former military. And so, they've already served several years and that's probably where their baseline even had existed. And then they've already seen the exposures of their military career and whatever outcomes of that then going back through the pipeline and getting back in shape. So, I agree. We're probably not catching the full longevity of these individuals. That's great. Or well said. Dr. Kaz. Yeah, I have a question. Is there any gym availability for the exercise in the fire station? Because there is, I deal with a lot of Montgomery County, Maryland firefighters and there's some stations have great gyms. And there was one time I asked one of the, in West Virginia, they don't actually allow exercise on site because they are afraid of the injuries. And so, this culture maybe, you know, of exercise is more important from station to station or from state to state. That's a good question. So, it varied, as you alluded to. From some of the stations did have a developed gym that was built out and some did not. And that may, you know, be a one item into our culture. But as we find a lot in society that simply owning or having something doesn't unfortunately translate to, you know, its use. It's a reality, right? You know, how many people buy gym sets or whatever or exercise bike and you have availability, it's there. Time's cut out for you. But there's something, some other component to that culture that, you know, environment that brings you to do that. You know, we watch people do, you know, a handful of pushups and, you know, then they were done, you know, after about 30 seconds. I think they, maybe they could have done more but weren't interested or something of that, you know, that being. But others, they were there, they saw each other and were kind of competing. So, it's definitely had a difference. OK. We'll take two more questions, one back here and then Dr. Hodge. First, yeah. Yeah, cool. Hey, Dr. Koh, good to see you. Hello. Quick question. So, why did you guys choose to use pushups as metrics and not doing like aerobic capacity testing or having them like do stair climbing in their bunker gear? Convenience. I think it ended up being just the simplest measure that we could do right there without, you know, significant risk of injury. It's, you know, there was a, we did cite a study that did find that that was a pretty good measure, not great. But it at least give us some idea of what their aerobic capacity is. But, you know, this again, one of our discussions is, well, we could do, you know, a run, we could do a stair climb, but to convince people to come back and participate in a more extensive physical fitness test was probably not as palatable as we wanted. Gotcha. Thank you. Thank you. And while we wait for Dr. Hodgson, does Mediterranean lifestyle contain like two glasses of wine and a couple hours on the beach each day? That was not a, that was not. It's just the diet, right. Disappointing. So, I just wanted to riff off of the exercise question. If you look at each of your risk factors or your components, you can define the social context in which that risk factor, that activity occurs. So, exercise, if you go to the gym, if you're a gym rat, you have a gym buddy, and it's the peer pressure of working with someone that will make you stick to something. If you stop smoking, I think the data are that 80% of couples that stop smoking at the same time stay quit. But if you don't have a partner who quits at the same time, it used to be about 15%. I mean, those are old data. I haven't done this kind of work in a long time. But you could go through cooking. So, Mediterranean lifestyle, even now, most firefighters are male. In general, on average, I was a volunteer firefighter in Connecticut there. Most of the meals were cooked by women. So, unless the spouse is invested in a Mediterranean diet, that's not going to happen. So, I wonder whether it's worth building into this project a broader culture survey that looks at each of the drivers for the things you care about as a success predictor. That's, and I think, you know, if you ran a research group with everybody here, you would come up with a model instrument that you could, you'd have to test out. But I think it would be actually not so hard to construct. That's a good point. Thank you. Great. Thank you, Dr. Koh. Thank you. Thanks again. All right, that concludes our session here. I invite you to buttonhole the speakers and ask them more questions. Please say thank you to a residency director or mentor of your choice, or all of them combined here. And we'll see you again in Philadelphia next year.

cold weather-related injuries

OSHA severe injury reports

trends

risk factors

awareness

prevention strategies

firefighters

initial training

BMI

body fat percentage

blood pressure

Mediterranean lifestyle