Dr. Tom Hamilton - Fatigue Management

And then a part of the EA team and, and, um, one of our attendings and has been, um, uh, gracious enough to help us through, uh, this fatigue management session, which is, um, uh, both required for us and, uh, gonna be helpful for us. So, uh, thank you very much. So don't worry, we're gonna all go through this pretest together. But uh his little background. Why is this important? No disclosures. So, the ACGMA or the Accreditation Council for Graduate Medical Education, uh, is important to all of us because once you graduate medical school in order to sit for your boards in any specialty, you have to go through an accredited institution. And the really overarching goals of this is to provide safe and quality care and try to make that as uniform as possible throughout its uh medical education programs. Every couple of years, they go through an extensive review, which is called this the CLR or the clinical Learning Environment Review. One of the major studies that undertook over 3 years where they visited 300 accredited institutions, interviewed over 9000 residents, 1000 faculty and leadership, and basically, We're most concerned about the safety in terms of how it related to not only work and duty hours, but fatigue and how that was managed in terms of education for the residents and fellows and the ongoing education of that for residents and fellows. So one of the things that we're gonna do today is simply prove to them that we have an ongoing Program where we're teaching you about fatigue and possibly ways to mitigate that or to uh make sure you're aware of these things and how they can impact patient safety. So, the good news for me is that Doctor Landrigan is an expert who is from Children's and has done a nice open pediatrics. So we're watching a video for the main portion of this. But we're just gonna go through the pre-test and this will help you sort of focus your attention during the video because you have to get an 80% on this test at the end. In order to uh get your credit. So we'll just go right through down and just mark it somewhere and then check it during the talk. Major drivers of human alertness and performance include acute sleep deprivation, chronic sleep deprivation, circadian misalignment, and sleep inertia. 2, sleep inertia is the tendency to continue sleeping when presented with an alerting stimulus when in slow wave, stage 3 or 4 sleep. The tendency to continue sleeping when presented with an alerting stimulus when in rapid eye movement sleep, cognitive sluggishness in the minutes to hours after awakening, that can increase the risk of error. The tendency of postoperative patients to have persistent drowsiness during the 24 to 48 hours following administration of common inhaled anesthetic agents. Number 3, 24 hours without sleep typically causes choose all that apply. Reaction time impairment equivalent to that induced by a blood alcohol level of 0.1%. Impaired judgment, loss of vigilance. Number 4. Motor vehicle crashes are the leading cause of death for individuals in their teens, twenties, and 30s. The best available data suggests that drowsy driving is responsible for what percentage of motor vehicle crashes. 35, 1020, or 50. Number 5, the likelihood that a resident physician driving home post-call will have a motor vehicle crash is about 20% higher than his or her risk when not post-call. True or false, Which of the following accurately reflect the literature on the 2003 Accreditation Council for Graduate Medical Education duty hour changes for resident physicians? The 2003 rules were associated with substantial reductions in sleep and improvements in mortality and complication rates. The 2003 rules were associated with substantial reductions in sleep, but little if any change in mortality and complication rates. The 2003 rules were associated with little change in sleep, but nevertheless improvements in mortality and complication. The 2003 rules are associated with little change in sleep and little change in mortality or complications. Effective strategies to fight drowsiness while driving include. All that apply, turning up the radio, chewing gum. Opening windows, pulling over and taking a 10 minute nap. A nap in the afternoon before a night shift must be at least 45 minutes in duration in order to improve alertness and performance in the night shift. So hopefully you'll have answers marked, and then we're gonna listen through this and then understand. OK, yeah, it's OK. I, I don't, there's not any more things that you have to worry about. Yeah. Ma Provider sleep deprivation in the ICU by Doctor Christopher Landrigan. Hi, this is Chris Landrigan. I am a pediatric hospitalist and patient safety researcher at Boston Children's Hospital. I'm also the director of sleep and patient safety program at Brigham and Women's Hospital and an associate professor of pediatrics and medicine at Harvard Medical School. I'm going to spend the next few. Worked 5 minutes ago. I'll get it. I log in again sort of thing. Hm, that's not supposed to happen. and countermeasures really this is about um the manner in which It doesn't mean you'll be 45. Provider sleep deprivation in the ICU by Doctor Christopher Landrigan. Hi, this is Chris Landrigan. I am a pediatric hospitalist and patient safety researcher at Boston Children's Hospital. I'm also the director of the Sleep and patient safety Program at Brigham and Women's Hospital and an associate professor of pediatrics and medicine at Harvard Medical School. I'm going to spend the next few minutes speaking about provider sleep deprivation in the intensive care unit, consequences and countermeasures. Really this is about the manner in which sleep deprivation and circadian misalignment can affect all of us when we're working in intensive care settings and some things that we can do to try to minimize its effects. So first of all, just a few disclosures. I have been supported in part by the Children's Hospital Association for my work on the PRI Research Network and I've also served as an expert witness and consultant on cases around sleep deprivation and patient safety, um, but otherwise no conflicts to disclose. Introduction and background. So I come at this by way of background in patient safety, and for me, really much of this begins with the perspective globally of how frequently things go wrong in hospitals. According to the Institute of Medicine's original report way back in 1999, it was estimated that between 44,000 and 98,000 patients are killed each year in the United States as a consequence of medical error. If we believe these numbers, it would make medical error something in the ballpark of the 6th and 9th leading cause of death in the country. More than AIDS, more than suicides, more than homicides, more than breast cancer. But despite this, it appears from more recent estimates that this continues to really be a major problem across the country. The Office of the Inspector General in 2010 estimated that perhaps there are as many as 180,000 deaths due to adverse events in hospitals, about half of which are preventable. And in a study that I led, which used a consistent methodology over time to try to track these trends and see if things were improving as time went by, we studied about 2300 hospitalizations in the state of North Carolina, and as you can see from the rather discouraging looking graph at the bottom of the slide there, we were not really able to see much evidence of improvement over time in that study either. Now this is not to say, by the way, that there have not been inroads in the patient safety movement in the past few years, for example, there has been tremendous progress in reducing the rate of nosocomial infections in the intensive care unit, for example, catheter-related bloodstream infections and others, and likewise, the rate of surgical complications in the operating room have improved tremendously in those hospitals that have begun to introduce standardized protocols and checklists and the like. As time has gone by, but these improvements have not yet translated up to the entire hospital level in large measure because of some fundamental underlying problems with our care systems that continue to persist. And that really permeate everything from the ICU to the ward setting to outpatient environment, and I would suggest to you that one of those, one of those forces is sleep deprivation and circadian misalignment. Alertness and performance. So if we look to the fields of sleep and circadian biology, there are really 4 major drivers of human alertness and performance that can affect the frequency with which errors are made in either hospital settings or elsewhere in the world, and these are biologic time of day or one circadian rhythm, the number of hours of consecutive wakefulness, nightly sleep duration, and lastly, a phenomenon called sleep inertia that I'll describe in a few minutes. In humans, the circadian rhythm is driven by a pacemaker that's located in the suprachiasmatic nucleus of the hypothalamus, and this pacemaker drives a host of secondary pacemakers around the body and in turn a diverse range of physiologic processes. Just to give you a couple of brief examples of this, most people are familiar with the notion, for example, that core body temperature follows a 24 hour pattern, getting a little lower in the middle of the night, a little bit higher by day. Likewise, plasma melatonin and growth hormone and cortisol and other hormonal outputs are driven, at least in part by a circadian rhythm. And in fact, if an individual is stuck in a cave-like environment without any external cues as to what time of day it is whatsoever. without any social cues as to what time of day it is. Nevertheless, these patterns will persist, so that about once per 24 hours, for example, the melatonin rhythm pulses, very much analogous to the manner in which the heart beats about once per second, but in this case this endogenous rhythm is once per 24 hours. From a safety standpoint, what's important about these rhythms is that they also drive alertness and performance. So we know, for example, that the frequency of slow rolling eye movements, the tendency for the eyes to begin to roll around in the sockets and for the head to nod, is much more likely to happen in the middle of the biologic night than the middle of the day. Likewise, things like reaction time begin to deteriorate very rapidly in the middle of the night. As you can see from this slide below, the temporal distribution of fatigue-related single vehicle trucking accidents tremendously increases in the middle of the circadian night at about 3:00 to 7 o'clock in the morning for most of us, suggesting that these types of biologic rhythms have important consequences in the real world as well as in our own bodies. Independent of the circadian rhythm is a second system called the sleep homeostat that's really a see-saw system that tells the body the longer that you've been awake, the greater the drive to fall asleep, and the longer that you've been asleep, the greater the drive to wake up. Again, from the trucking literature, you can see the relationship between this force and the propensity for motor vehicle crashes to occur. In this case, this is displayed not by time of day but by a number of consecutive hours driving. And what you're seeing is that the risk of a fatigue-related motor vehicle crash in truckers is relatively flat until about 7 or 8 hours of consecutive driving, at which point you begin to see an exponential rise such that at the 13th hour of consecutive driving, the risk of a fatigue-related crash is 1,500% higher or 15fold higher than at baseline. The amount of impairment that's induced by acute sleep deprivation has in fact been very consistently shown to mirror that which is induced by alcohol, so that at about 17 to 19 hours of sustained wakefulness, depending on the study, performance impairment is equivalent to that induced by a blood alcohol level of about 0.05%. And at the 24 hour mark, on average, our performance drops to that which is induced by a blood alcohol concentration of 0.1, which is beyond the legal limit in the United States. In the past 20 years, it's also been recognized that sleep homeostatic drive is mediated not only by the amount of sleep obtained in the prior 24 hours, but also by the amount of sleep that's being obtained on a regular basis over the preceding days to weeks. In this experiment, what you are seeing is a group of research subjects who have been restricted to 8 hours' time in bed, 6 hours' time in bed, 4 hours' time in bed, and their performance is being compared with a group that has been restricted completely, so no sleep whatsoever for 24, 48, 72 hours in the black line all the way to the left of the figure. And what you're seeing is that the 6 hour time in bed group, for example, if you focus in on the yellow line and the yellow dots, by about 10 to 14 days of 6 hours time in bed with no chance for recovery sleep begins to perform at a level that's similar to that that's induced by 24 hours of total sleep deprivation. By comparison, the 4 hour time in bed group by 7 to 10 days of chronic sleep restriction is performing like the group that's had no sleep whatsoever for 48 consecutive hours, which is really a pretty profound level of impairment. And as you can see from the 2nd figure on the right, an interesting fact about chronic partial sleep deprivation is that our ability to perceive that this is happening to us as it's taking place begins to degrade as time goes by. So focusing in on again just the red line for a moment, you can see that in the first couple of days subjective rating of sleepiness more or less mirrors the slope of the objective line over in the left hand figure, but as time goes by, you see a plateauing, which is suggesting that folks are saying as time goes by, maybe I'm a little bit more tired than I was last night, but it's not so bad, and I'm beginning to get used to it. I'm beginning to compensate for this type of schedule, whereas in fact, objectively what we're seeing is things are getting worse and worse and worse in a fairly linear fashion. Um, even as the same individuals are saying that things are fairly stable and not getting too much worse. 1/4 factor just to mention very briefly is a phenomenon called sleep inertia, which is really the notion that the brain does not go from 0 to 60 in 2.5 seconds. In fact, for the first, especially 5 to 15 minutes, but with some residual effects lasting out to as long as 2 hours, there are impairments in certain areas of cognition, particularly the prefrontal cortex and the thalamus and elsewhere in the brain, which can greatly increase the risk of error and injury in those first few minutes. This was interestingly first discovered in the US Air Force and then rediscovered in the Israeli Air Force back in the 1980s where at that time pilots who were in potentially hostile situations were sleeping in their cockpits ready to be scrambled for duty and the sirens would go off. These fighter pilots would go down the runway and they were promptly crashing into the mountains and crashing into the ocean at a very high rate, much, much higher than they were at baseline. At which point the Israeli Air Force, as the US Air Force had before, got rid of the policy of having pilots sleep in the cockpits. Experimentally this has been replicated a number of times since then and particularly in the first, let's say 5 minutes after awakening from deeper stages of sleep, stage 3, stage 4 sleep, the amount of impairment that's induced by sleep inertia can exceed that which is induced by 24 hours of acute total sleep deprivation, profound disorientation, and a very, very high risk of error in those first few minutes after awakening. Consequences for healthcare providers. As healthcare providers, of course we're exposed to all four of these factors on a regular basis. Very frequently it's the case that we're working in the middle of the biologic night, often in the setting of having not gotten good sleep over the prior 24 hours, perhaps 1 hour, perhaps 2 hours, perhaps not at all. This may be in the background of not having gotten adequate sleep over the prior weeks for both personal and professional reasons. And if you are so fortunate as to sleep for a few minutes in the hospital and are suddenly awoken for a code blue or another type of an emergent situation, then sleep inertia can really come into play as well. Now, the risk of sleep deprivation impacting physicians has been a source of study for over 40 years. Now the very first study that looked at this issue came out in the New England Journal of Medicine way back in 1971, where a group of interns were asked to read electrocardiograms when sleep deprived versus rested, and it was found that they made about twice as many mistakes when reading reading those studies when sleep deprived. In this meta-analysis, Ingrid Philibert from the ACGME put together a whole series of these studies, about 60 in total. Some that included physicians, some that included other individuals, and tried to look across a range of different subtly different conditions, what the effects of sleep deprivation were on performance. And as you see from the figure on this slide, this is sort of a typical meta-analytic figure where the zero effect line is that thick dark gray bar, vertical bar in the middle. Any study that crosses that bar, the 95% confidence intervals crossed that line, would come to the conclusion that sleep deprivation has no effect on performance. And you can see that there are some studies that have reached that conclusion, but if you look at the aggregate message from all the studies combined, it's pretty clear that things are on one side of that bar, namely to the left hand side, indicating that the sleep deprived group did worse than their rested peers. What this suggests is that. First of all, consistently across a range of different types of studies that have been conducted, at least in protected laboratory environments, sleep deprivation really profoundly affects the performance of physicians because what we're seeing here is not just a mild decrease in performance, but on average a 1.5 standard deviation drop in performance with 24 hours of acute sleep deprivation here. And if we were to focus in on the subset of these studies that have looked at more clinical tasks such as reading EKGs or doing simulated surgery. In fact, it's about a 2 standard deviation drop in performance to the 7th percentile of rested baseline performance, which I think is not where any of us would like to be when performing medicine. But a very reasonable criticism of this work up until about a decade or so ago is that for the most part it was really conducted in controlled laboratory types of environments. And so it may be the case, for example, that sleep deprivation adversely. Affects the ability of an intern in a dark room after a long shift to read an EKG, but how does this play out in actual clinical practice? In other words, if the adrenaline is pumping and a patient is having a code and and and things are much more exciting in the real world, are the same types of effects from sleep deprivation still going to be seen? And so what my group set out to do over a decade ago now was to study this issue, and we did that really in two ways principally. The first is that we conducted a study, a national cohort study of about 3000 interns nationwide beginning in 2002, 2003. Where on a monthly basis we had these interns report for us exactly how much they were sleeping, how much, how much they were awake, um, how much they were working, and then any time they had a car crash or a needle stick injury or thought they made a medical error, we had them send us reports of those and we validated them in various ways. What we found, I'm really very commensurate with the trucking literature in some respects, is that the odds that an intern would have a motor vehicle crash on the drive home from work was more than doubled when driving home post call after a 24 hour shift as compared with him or herself on other days of the week. This was a so-called within subjects type of a study where individuals were compared with themselves in different conditions, in other words, post-call versus not post call in this case. We also found that the risk of a needle stick injury or a scalpel injury for a surgeon or obstetrician was about 60% higher after 24 hours of work or near 24 hours of work as compared with that same individual on other days of the week when they had not been working an extended shift. In addition, at least by the intern's own reports, we found that 1 in every 5 interns reported making a fatigue-related mistake that injured a patient at some point over the course of their internship. And 1 in every 20 reported having made a fatigue-related mistake that resulted in the patient's death. These were very strongly correlated with the number of overnight shifts that were worked in a month. And if we were to extrapolate this out to a national level just to give you a sense of the scope of these reports, this would suggest that something in the ballpark of 5000 to 10,000 deaths due to intern fatigue or resident fatigue occur each year in the United States. That said, there are reasons to be unsure about any self-reported measures such as this. It's not clear that the scope of things is that large. However, it's not clear exactly what the true number is. And so we, um, at the same time that we were conducting this national cohort study, we sought to more objectively measure the frequency with which adverse patient safety events occurred in the context of ICU care by conducting a randomized controlled trial and really what we did in this initial experiment back at Brigham and Women's Hospital about 10 years ago now. was to randomize the interns to work a traditional schedule for one of their two ICU rotations during that year, which was a Q3 schedule relying on repetitive 30 hour shifts, and they were compared with performance on an intervention schedule where we had 4 residents working a schedule where they rotated through days and nights and never worked for more than 16 scheduled hours in a row. What we found in doing this is that despite having bumped into problems with handovers and some difficulty making the education for those residents work smoothly, despite that, the traditional schedule appeared to be far less safe than the intervention schedule. The interns overall made 36% more serious errors in the care of their patients, including about 20% more serious medication errors and more than 500% as many serious diagnostic errors, which included everything from Failures of vigilance, forgetting to follow up on laboratory values or misreading chest X-rays to the much more complex task of putting together all of the bits of information and the history of physical and lab results and so forth to come to an appropriate conclusion about what to do with the patient. Now one of the concerns about the potential of reducing. Hours in intensive care units or for that matter in other settings in medicine is that if we were to reduce the hours of trainees, we would also have a problem with those individuals not having sufficient exposure to case volume and so forth to be adequately trained at the end of their programs. However, while this is certainly a reasonable concern, a countervailing concern is we also know that individuals who are sleep deprived have a very difficult time learning effectively and in particular consolidating memories that are obtained. What you're seeing in the slide here is that. Individuals who are presented with a novel task, this is a laboratory type of learning experiment, a novel task on day zero are asked to do a visual discrimination task. You can think of it as it's it's, it's a game similar to the old Tetris video game, if you will, where shapes are being fit together that are presented in random order, something that these subjects would never have had a chance to see before. And what you're seeing on the slide is that if individuals are presented with this task on day 0 and then some of them are brought back on 1, some on day 2, some on day 3, day 4, day 7, even with no opportunity to practice this novel task in the interim, they're better at it when they've come back to the next day or even several days later. However, if they are sleep deprived the night after having learned this novel task, what we find is that. They have experienced no learning or almost no learning whatsoever over the intervening days, indicating that there's something that's crucial about sleep itself that consolidates these memories and that allows them to subsequently improve even in the absence of practice. Now, how this plays out in the real world of medical education is a little bit unclear, but it may be the case that performing more, uh, let's say, appendectomies if you're a surgical trainee when sleep deprived may or may not be better than less of them when when well rested. If we look at the larger literature on the effects of reducing resident work shifts, in particular of eliminating shifts of greater than 16 hours, studies that have been done in large part since our randomized controlled trial of a number of years ago. Most of these studies have suggested, first of all, that resident quality of life improves. Secondly, the effects of these types of work hour changes across those programs that have tried them on resident education are surprisingly neutral. 9 out of the 14 studies that have looked at these types of outcomes to date have not found that resident education has been adversely affected by reducing hours. In fact, there's some slight tendency towards improvement in learning when hours are reduced. And from a safety researcher's standpoint, at least, most importantly, 7 of the 11 studies that have looked at outcomes of safety, quality of care, and so forth have found that safety and quality improve significantly when hours are reduced. 4 of the studies have shown that there's been no change. It's been a wash. There are no studies to date that have demonstrated that making these types of changes leads to a reduction in patient safety. Policies. Now from a policy perspective, in light of these literature and a range of similar studies that have come out of other industries, the Accreditation Council for Graduate Medical Education began to feel in the early 2000s that something should be done about resident work hours and for the first time nationally at that point in time implemented work hour limits for residents, and those limits were no more than 80 hours per week, averaged over 4 weeks, no more than 30 hour shifts, and one day off in 7 again averaged over 4 weeks. Um, unfortunately, at that time, the limits that were put in place were really relatively modest in many respects and did not lead to very substantial reductions in most medicine, family practice, and pediatrics programs. And as a consequence, not surprisingly, the data on the effects of these types of policy interventions on real outcomes for patients have been relatively modest as well. There have been a couple of studies that have suggested that perhaps there was a mild reduction in mortality or complications for some medical patients, but the largest studies conducted really found no effect either on surgical patients or medical patients after this policy change went into effect. However, again, when we look at the actual change in work and sleep hours, this is again from our national cohort study, we found that there's only about a 5% reduction in actual work hours, only a 5 or 6% increase in weekly sleep hours, and importantly, in the bar graph at the far right of this slide, you can see that there was actually a hair reduction in the amount of sleep that was being obtained during extended duty shifts when this policy went into effect. And since it's the amount of sleep that's being obtained during extended duty shifts we think is the major driver of sleep-related errors in hospitals, if we haven't made any progress on that measure, then we really haven't made any progress at all. And thus again it's not surprising that outcomes did not appear to improve very much. To take a bit of a step back, it's also important to recognize that while more studies have been conducted in medicine in resident population than in populations of others such as nurses or attending physicians. Those studies that have been conducted in nurses suggest much the same thing as the studies of residents. So Anne Rogers and colleagues published a decade ago now that shifts of more than 12.5 hours were associated with the tripling of reported medical errors among nurses. A very similar study was conducted by Scott and colleagues among critical care nurses in particular, where a doubling in error rate was found. And uh nurse needle stick injury risk as well as driving risk in subsequent studies has also been found to increase when nursing shifts go above about 12.5 or 13 hours. For attending physicians in particular, while the literature is a little bit more mixed here and far more preliminary in general. We conducted a 10 year retrospective study of attending surgeons and obstetricians at Brigham and Women's Hospital a few years back where we found that the rates of complication in post-nighttime procedures were not much different overall than those in control nights. But when we looked more deeply at the amount of sleep that was obtained during on-call shifts in particular, when the surgeons obtained less than a 6 hour opportunity to sleep at night, the odds of complications in the operating room the next day nearly tripled. In light of these data, the Institute of Medicine in 2008 to 2009 convened a group of individuals really from across specialties and expertises to come together and to make recommendations on what should be changed about resident work hours. And really after reviewing the world's literature on this topic and holding a whole series of hearings, the panel concluded that fundamentally it was unsafe for residents to be working for more than 16 hours in a row without sleep. They proposed a couple of possible solutions the implementation of a mandatory sleep period or nap period in the midst of a 30 hour shift, or alternatively, a reduction in shift lengths to no more than 16 consecutive hours. Recognizing that fixing work hours all by itself is not going to fix anything. However, they also called for an improved handoff processes, improved supervision of residents, and really a fundamental restructuring of the infrastructure of residency programs in order to facilitate both shorter hours and safer care overall. Responding both to these recommendations as well as to its own internal reviews, the Accreditation Council of Graduate Medical Education in 2011 implemented a 16 hour shift limit for interns but continues to allow PGY 2 and higher residents to work for up to 28 hours in a row, in large measure really based on the fact that most studies have been conducted on interns to date and the literature with respect to the more senior trainees is a little bit less robust to date. Several studies have come out within the past year or so looking preliminarily at the effects of these 2011 policy changes. For the most part they've really been both mixed and non-effinitive, I think would be the right way to put it. Desai, for example, and colleagues in JAMA Internal Medicine in 2013 found that sleep did indeed improve with implementation of the 2011 standards in that program, but the number of handoffs also increased and educational opportunities decreased. By contrast, at Vanderbilt University, Theobald and colleagues found that inpatient encounters and attendance at educational conferences improved in the wake of the duty hour changes, suggesting that perhaps the manner in which these changes were made at the two institutions might have very substantially affected its effectiveness. In a national study of self-reported errors by residents, Sen and colleagues found a slight increase in the number of self-reported errors that were reported by the residents themselves, but whether this in fact represents an increase in errors in the wake of the policy change or simply an increased sensitivity to errors in the wake of the change remains unclear, as objective data have yet to be collected. Summary and future steps. Overall, field and lab studies across disciplines show consistent performance decline after 12 to 16 consecutive hours of work. We know that chronic sleep deprivation, circadian misalignment can play into this as well, such that critical zones of vulnerability exist when all three of these factors are present together, and the risk of errors, at least in laboratory settings, can increase very, very substantially. We know from a series of studies among residents themselves that eliminating 24 hour shifts can reduce medical errors. Furthermore, we know that residents working 24 to 30 hour shifts have twice the odds of crashing their cars, which is the number one cause of death in people in this age group, and they suffer a 61% higher risk of percutaneous injuries. Overall, they perform at a level commensurate with that induced by a blood alcohol of 0.05 to 0.1, which equates to a 1.5 to 2 standard deviation drop in performance across medical and non-medical tasks. And we know that improving sleep improves learning and memory, although certainly decreased hours in the hospital could have a countervailing effect. When thinking through how we can take these data and put them into practice in hospitals, there are several ideas that emerge. The first is limit shift length to no more than 12 to 16 consecutive hours, um, similar to what the Institute of Medicine recommended and to what the ACGME itself has implemented for interns, if not for more senior residents. Minimizing consecutive nights on duty is also important, particularly from the occupational literature in mining and factory workers and so forth. We know that working more than 3 or 4 consecutive nights in a row tends to increase error risk as well. And so a number of schedules have emerged that would suggest that perhaps one solution to this is so-called rapid cycle rotation, where someone who has to work nights in a 24 hour industry only does 1 or perhaps 2 in a row before going back to a day shift. We know that providing sufficient time off between shifts is critical, a minimum of 10 hours in order to allow for about 8 hours of actual sleep time. Providing adequate time off for recovery after a string of nights is also important. And in so doing, matching staff needs to workload, really thinking carefully through at what time of day is staff needed to be in the institution and how can we make sure that that workload is aligned with those staffing needs is very important as well. One mistake I think that has been made in many institutions as they've begun to try to make these changes is that sometimes the workforce is thinned in such a way that those residents or other workers who remain are effectively overworked, and this can lead to a trade-off between fatigue related errors and errors that are related simply to being overworked and having too much to do, and that's something that obviously we'd like to avoid in trying to design solutions. In hospitals in particular, it's really critical as hours are reduced to make sure that there is a high quality handoff process, a change of shifts, so that there's not a loss of information, and again an increased risk of errors as a consequence of that. It's also really important to make sure that there's adequate supervision in place and to think very creatively about educational curricula. If we have an educational program that is really built around 24 hour shifts, but now not all of our residents are present for those 24 hour shifts and therefore are missing their critical educational. Opportunities then there needs to be a redesign of education such that um residents can can perhaps do some on their own, have more of a chance to sort of really maximize that time that they do have in the institution to get feedback from preceptors and more senior mentors and so forth. What can residents themselves do? Well, first of all, it's really important to sleep when you can. Um seems like a simple thing to do, but I think all of us in medicine tend to underestimate the importance of sleep. It's critical to try, particularly if sleeping during the day, to optimize sleep hygiene, the notion that you should really be sleeping in a cool environment. For most people, about 60 to 65 °F is optimal for best consolidated sleep. It should be truly dark. Blackout shades are a must. If trying to sleep during the day, it should be quiet to the extent possible. Try to avoid having any beepers or phones or other possibilities for interruptions in there during a period of sleep. The use of white noise machines or fans, which can both cool the room and provide white noise, can be very, very valuable. Um, overall it's, it's important for residents to recognize and for others to recognize that sleep is a priority, but it's also a responsibility. Um, a schedule can provide opportunities to sleep, but it's really only the individual that can take advantage of that opportunity. Napping before night shifts is one way to optimize overnight performance, a very important way. We know, for example, that even a 20 minute nap in the afternoon decreases the tendency for performance to fall to its nadir during the during the subsequent night on a night shift. Most of us will get our best opportunities to nap between about 3 o'clock and 5 o'clock in the afternoon just based on the way that the circadian system works for most of us. Thirdly, if you have the opportunity to go home after a, after a night shift, it's best to do so as soon as you can. Sleep that's obtained beginning as early in the morning as possible after a night shift will tend to be of higher quality. And of longer duration than sleep that is delayed for a couple of hours. So in other words, if you get to get off shift at 7 o'clock, much, much better to go home and sleep at 8 o'clock than to run a few errands and try to go to sleep at 10:00 a.m. You'll get much, much better sleep if you try to go to sleep right away and put off those errands and so forth until until later in the day after you wake up. And then remember that while on shift, the caffeine works. Caffeine is very effective as an alertness promoting agent. It's most effective if taken in small quantities throughout the shift. So a gigantic coffee taken at the beginning of the shift tends to be much less effective than small amounts of caffeine, even a quarter to 1/2 of a cup taken on an hourly basis over the course of the shift. Lastly, um, very important to try to avoid drowsy driving as much as possible. Drowsy driving is responsible for as much as 20% of motor vehicle crashes overall, which is in turn the leading cause of death for people in their twenties and thirties. To the extent that a taxi service is available, which is a possibility in many institutions, it should be used, use public transportation whenever possible. Try to minimize commute distance whenever that's a possibility as well, and in particular avoid traveling at high speeds on highways and parkways where sleep propensity tends to be highest. And most importantly, if you do feel sleepy when driving, if your eyes start to close and your head starts to nod, the only thing that's effective is to pull over and to take a nap, then to allow yourself a few minutes to wake up and get through the worst stages of sleep inertia and then drive back home, rolling down the window, turning up the radio do not work. Chewing gum does not work. The only thing that works is to take a nap. Um, with that, I'd like to stop and thank my collaborators in the Harvard Work Hours Health and Safety Group as well as the Center of Excellence and Patient Safety at Brigham and Women's, um, Boston Children's, Brigham and Women's and Harvard Medical School, and all of our funders over the years. Thanks very much. That concludes our video on provider sleep deprivation in the ICU. Please help us improve the content by providing us with OK, so hopefully you'll absorb that, and we're not too tired to uh get it in. So the good news is if you scanned your badge, you will get credit, but we're gonna go over these questions uh together to see if we were paying attention. OK, so number one, major drivers of human alertness and performance include Choose all that apply acute sleep deprivation. Yes. Chronic sleep deprivation? Yes. Circadian misalignment? Yes. Sleep inertia. Yes. What is sleep inertia? The tendency to continue sleeping even when presented with an alerting stimulus when in slow wave 3 or 4 sleep. No takers. The tendency to continue sleeping even when presented with an alert alerting stimulus when in rapid eye movement, REM sleep. No takers. Cognitive sluggishness in the minutes to hours after awakening that can increase the risk of error. A lot of takers. Correct. 24 hours without sleep typically causes, uh, choose all that apply, reaction time impairment equivalent to that induced by a blood alcohol level of 0.1. Correct. Number 4. Motor vehicle crashes are the leading cause of death for individuals in their teens, 20s, and 30s. The best available data suggests that drowsy driving is responsible for what percentage of motor vehicle crashes. 3 5 10? 20? For 50. 20 is right. The likelihood that a resident physician driving home post-call will have a motor vehicle crash is 20% higher than his or her risk when not post-call. True or false? It's true. Twice. Any takers, true or false? Higher than 20, 50%. Which of the following accurately reflect the literature on the 2003 Accreditation Council for Graduate Medical Education duty hour changes for resident physicians? The 2003 rules were associated with substantial reductions in sleep and improvements in mortality and complication rates. The 2003 rules were associated with substantial reductions of sleep, but little if any change in mortality and complication rates. Any takers? Not really. The 2003 rules were associated with little change in sleep, but nevertheless improvements in mortality and complication rates. Or finally, 2003 rules were associated with little change in sleep and little change, if any, in mortality and complications. Takers? Correct. Effective strategies to fight drowsiness while driving include turning up the radio. Turn up the bass chewing gum, opening the windows, or pulling over and taking a 10 minute nap. Take the nap, but also wait 15 minutes so you're cognitively not impaired. A nap in the afternoon before a night shift must be at least 45 minutes in duration. 2020 was the magic number. So that concludes our session to fight your fatigue. We'll let you out a little few minutes early. And hopefully, you've learned something about how to take care of your own sleep today. Thank you. It Well, I Good. I know. long Yeah. Yeah I I. I. Yeah. I OK. But He's not. OK Thank. Yeah. yeah. Right, right, right, they've already Yeah You I Yeah. I get better after you get the line up. That's my problems. I My, my grandma, yeah, they, they had. you Perfect. Maybe I'll have days have someone else do it.