National Sleep Awareness Week might have been last week, but many of us are feeling the importance of shuteye this week, as we struggle to drag ourselves out of bed at what feels like an inappropriate hour. While Daylight Saving Time may get the blame for sleepiness this week, though, there are important year-round factors that cause fatigue. In honor of National Sleep Awareness Week, the National Institute for Occupational Safety and Health’s NIOSH Science Blog published two posts about the impact of work demands on sleep. Claire Caruso and Roger Rosa start off by highlighting the challenges of long hours and shift work:
The timing of a shift can strain a worker’s ability to get enough sleep. Working at night or during irregular hours goes against the human body’s biology, which is hard-wired to sleep during the night and be awake and active during the day. Still, society needs certain workers around the clock to provide vital services in public safety, healthcare, utilities, food services, manufacturing, transportation, and others. The resulting shift work–any shift outside the normal daylight hours of 7 a.m. to 6 p.m.–is linked to poorer sleep, circadian rhythm disturbances, and strains on family and social life. It is not possible to eliminate shift work altogether, so the challenge is to develop strategies to make critical services available while keeping workers healthy and everyone around them safe. In addition to shift work, some data suggest that a growing number of employees are being asked to work long hours on a regular basis. Every extra hour on the job is one less spent attending to the person’s off-the-job responsibilities. When the day is too full to fit everything in, it is often sleep that gets the short shrift.
They list some of the risks to workers from long work hours and shift work, including a decline in mental function and physical ability; decline in immune-system functioning; higher rates of depression; an increased risk of heart disease; and an increased risk of illness or injury. These risks to workers also matter for employers, who can see reduced productivity, more errors, and increased healthcare and workers’ compensation costs. And, of course, fatigued workers can make medical errors and contribute to car crashes and industrial disasters whose impacts extend far into their communities.
In an American Journal of Public Health article offering recommendations for shift duration and sequence during emergency response activations, Paula Burgess focuses on the times of day at which industrial and engineering disasters have occurred, suggesting “a possible association between human error and circadian rhythm as it relates to shift work.” She gives several examples:
The incident at the Three Mile Island nuclear facility in Pennsylvania on March 28, 1979, occurred because night shift workers failed to recognize the loss of core coolant water resulting from a stuck valve between 4:00 AM and 6:00 AM. Similarly, on June 9, 1985, the Davis-Besse reactor in Oak Harbor, Ohio, went into automatic shutdown followed by a total loss on the main feed water at 1:35 AM. The incident reached more critical proportions when an operator in those early morning night shift hours pushed the wrong 2 buttons in the control room, defeating the safety function of the auxiliary feedwater system. When the Racho Seco nuclear reactor near Sacramento, Calif, automatically tripped after DC power to the integrated control system was lost at 4:14 AM on December 26, 1985, human errors of omission and commission caused time delays in regaining control of the plant. Even the nuclear plant catastrophe at Chernobyl is officially acknowledged to have begun at 1:23 AM as the result of human error.
The Presidential Commission on the space shuttle Challenger accident cited the contribution of human error and poor judgment related to sleep loss and shift work during the early morning hours. This same commission also cited early morning shift work error in a previous near-catastrophic launch of the shuttle Columbia on January 6, 1986.
The incidents involving the Bhopal Union Carbide tragedy and the Exxon Valdez, as well as the Estonia ferry incident, all occurred in the early morning hours. Investigations have concluded that all of these were at least in part attributable to fatigue and human error.
Burgess recommends that when 24-hour emergency coverage is necessary, 8-hour shifts are preferable to 12-hour shifts, and if the size of the personnel pool allows it, shifts for each worker should rotate as follows: three day shifts, followed by three evening shifts, three night shifts, and three recuperative days off.
Caruso and Rosa also have suggestions that apply to all employers and workers, not just those engaged in emergency response:
What can employers do to address this issue?
- Regular Rest: Establish at least 10 consecutive hours per day of protected time off-duty in order for workers to obtain 7-8 hours of sleep.
- Rest Breaks: Frequent brief rest breaks (e.g., every 1-2 hours) during demanding work are more effective against fatigue than a few longer breaks. Allow longer breaks for meals.
- Shift Lengths: Five 8-hour shifts or four 10-hour shifts per week are usually tolerable. Depending on the workload, twelve-hour days may be tolerable with more frequent interspersed rest days. Shorter shifts (e.g., 8 hours), during the evening and night, are better tolerated than longer shifts.
- Workload: Examine work demands with respect to shift length. Twelve-hour shifts are more tolerable for “lighter” tasks (e.g., desk work).
- Rest Days: Plan one or two full days of rest to follow five consecutive 8-hour shifts or four 10-hour shifts. Consider two rest days after three consecutive 12-hour shifts.
- Training: Provide training to make sure that workers are aware of the ups and downs of shiftwork and that they know what resources are available to them to help with any difficulties they are having with the work schedule.
- Incident Analysis: Examine near misses and incidents to determine the role, if any, of fatigue as a root cause or contributing cause to the incident.
What can workers do to address this issue?
- Make sure you give yourself enough time to sleep after working your shift.
- Avoid heavy foods and alcohol before sleeping and reduce intake of caffeine and other stimulants several hours beforehand since these can make it difficult to get quality sleep.
- Exercise routinely, as keeping physically fit can help you manage stress, stay healthy, and improve your sleep.
- Choose to sleep someplace dark, comfortable, quiet, and cool so you can fall asleep quickly and stay asleep.
- Seek assistance from an appropriate healthcare provider if you are having difficulties sleeping.
Also, if you find yourself staying up later than you should clicking on one YouTube video after another, remind yourself that sleep is important for safety — then turn off the computer and get some sleep.
Isn’t this based on a false assumption that workers should always be well rested or that such systems are affordable? There are costs to overworking people, true, and fatigue can cause accidents, but it’s unlikely we will adopt such lax hours in an industry like healthcare because we simply couldn’t afford it. If I work 80 hours a week with 1-2 30 hours shifts interspersed between a few 10 hour shifts (which does have continuity of care benefits) I might not be 100% sharp all night, but to replace me under this regimen you’d have to have 2 or 3 other workers. While I would love to work 8 hour shifts and get “light” duty (haha) if I had to work 12, there is simply no economical way to make this feasible for my industry. You’d have to hire 3 times as many residents.
Instead these reports lead to the medical equivalent of the unfunded mandate. We are forced to obey these hours schemes while the ACGME and federal government limit our total number of residency spots. This leads to worse patient care because in order to keep everybody out of the hospital, we have to take 3 times the work, do it in a shorter period, and then when on call, cover 3 times as many services. More time while at work gets devoted to scut and paperwork, and we get less time performing tasks that are critical for the training aspect of residency, like OR time and clinical decision making.
I guess what I’m saying is, the work isn’t going to go away, so how are we going to pay for this? Yes in a perfect world we’d be well rested all the time and teacher would give us nap time after a big meal, but we don’t live in that world. We’d be well rested but such a system would cost 2-3x as much and require 2-3x more people. And please don’t suggest a mandate for restrictions to be implemented without legitimate plans for workforce expansion. It makes our lives harder, not easier.
From what my night-shift ICU-nurse friends have described, sometimes the biggest barrier to good sleep isn’t something the employer does so much as the fact that most of the rest of society doesn’t understand shift work, and so will call you up, ring your doorbell, run the lawnmower, etc in the middle of your sleep period. I try hard to use non-invasive communication methods (e-mail), but the fact is that you can’t tell everyone, and lots of people think if you’re home during the day you are free to bother.
(I’ve also had a few arguments with bosses about how you don’t get good science out of someone who’s been up for 20+hrs.)
Heh. Same here in live theater — my main work hours are nights and weekends. And inevitably, people will want me at 8:00 AM Monday…and then proceed to snicker “Oh, you artists — you do love to sleep in, don’t you!”
Mark, you’re right that a switch to sleep-optimal schedules in healthcare — in particular, hospitals — would be unaffordable. I don’t expect hospitals to be able to comply with all of NIOSH’s recommendations, although I think it’s useful to bear in mind that they’re the ideal and see if it’s feasible to make small steps toward them (e.g., how break time is distributed during a shift).
You also question the assumption that workers should always be well rested. There are two issues here: first is whether circadian disruptions and insufficient sleep have adverse effects on cognitive performance, and there’s compelling evidence that they do. The second is whether there are sufficient safeguards in place to prevent cognitive deficits from leading to injuries or medical errors. Hospitals have multiple providers involved in caring for the same patients, so a lapse on the part of one provider might be caught by another — or an electronic ordering/prescribing system might be able to flag potential adverse outcomes.
You also bring up the issue of whether long shifts have benefits that would be reduced under shorter shifts, and that’s also something hospitals have to consider — understanding, as Darshak Sanghavi points out, the importance of handoff quality in comparing short shifts to long ones. It’s entirely possible that the cost-benefit profile of sticking with long shifts is better than switching to short ones given the current patient volume and financial realities. It’s important, though, to be clear about what the costs of the current system are.
I’d be curious to know from your (and others’ ) experience whether hospitals explicitly acknowledge the cognitive effects of sub-optimal shift scheduling and take steps to reduce the risk of injuries and adverse events that can result. Are there safeguards or trainings related to this? Is there a culture that allows for providers to ask for something — scheduling changes, a place to sleep, cab fare home, extra help during the last two hours of a shift, etc. — to mitigate the effects of less-than-ideal shift work?
Oh, and I agree that unfunded mandates are a problem. I’d like to see the federal government provide more funding for residency slots, anyway — in particular, slots for training primary care doctors — but this isn’t a great time to be asking for more spending on healthcare.
I participated in a conference last week with hundreds of industrial workers from plants all across the U.S. My jaw dropped when I heard them describe their typical weekly schedule of 65-75 hours per week, with shifts changing from days to nights to afternoons. Yes, they get lots of overtime pay, but everyone I heard from said they would easily give-up the additional pay for a 40-hour work week and a predictable schedule. It seems to me that some of our unemployment problems could be solved if employers would hire more workers.
We get mandatory sleep training at the beginning of the year. But really during training we learning coping mechanisms to prevent error. An effective resident organizes things in such a way that mistakes don’t occur. You make a list for everything, you check it twice, you organize your day so that critical tasks become routine. You round before it get’s too late and try to take care of things while you’re still fresh. And since medicine is a practice, eventually you get to the point where you don’t even have to think anymore.
Being tired is one thing, but what’s really dangerous is being tired and disorganized.
Sounds sorta dangerous, when, if ever, I am (or anyone) and the attending is organized, but his or her brain is off, they just MIGHT miss something, something they should have caught, but was outside of the cookie cutter organized, but sleep deprived MD/DO.
And yes, the few tax payers that are left in this country are not going to want to send more dollars to the healthcare money pit.
DavidO, read up on the science of expertise. It’s not that your brain is “off”, but with practice decisions of experts become lightning fast and no longer require you to “think” about it like less experienced personnel. Rather than making decisions based on a weighing an excess of options and data, the expert becomes a pattern-recognizer. You don’t have to think about things anymore, you just know what to do. It’s counter-intuitive, but true.
I see it in the way I practice. I don’t need to weigh every lab and search the literature and obsess over every abnormality in the exam like I did as a medical student. With experience you have to think less, you recognize patterns, you learn a routine which brings you to the right answers and the right outcomes, and you see in the less experienced they’ll get to the same answer much of the time, but it will take them 10x longer.
It seems to me that some of our unemployment problems could be solved if employers would hire more workers.