The Truth About Split Shifts

Author:  FMN Contributing Author
Split Shifts

The problem with split shifts is that they can cause unpredictable sleep that can lead to fatigue.  Sometimes these shifts lead to split sleep, sometimes they lead to short sleep and sometimes they have no impact on sleep, and shift-workers don’t like them.

If your split shifts occur during the day, and still provide 12 consecutive hours off with the ability to sleep for at least 8 straight hours somewhere in between about 22:00 and 09:00, there will probably be no impact on your shift-workers’ sleep and no increase in fatigue risk specifically due to the split shifts.

If your split shifts encroach into the nighttime or early morning hours and you are hoping that your shift-workers will nap in between the split shifts to make up for working during sleep time, you might be inadvertently increasing fatigue risk.  It is very difficult to predict who will nap, when they will nap and for how long they will nap.  It takes a very committed workforce, an ideal sleep and fatigue culture across the organization and advanced training on the science of napping to keep this risk at bay.  Without these factors, shift-workers will usually just try to survive on the short nighttime sleep required by the split shifts.  Short sleep increases the risk of fatigue. 

If, however, you manage to get your shift-workers to augment their night sleep with a nap, that is, they split their 8 hour sleep into 1 main nocturnal period and a daytime nap, their performance and fatigue should be satisfactory[1].  This is known as biphasic sleep.  The key here is that biphasic sleep can be ok only if the shift-workers’ main sleep periods are at night.  Biphasic sleep will not always be night sleep plus a nap.  It could be a main day sleep period and a nighttime nap.  With daytime sleep being the main sleep period, sleep quality is likely to be low and this means supplementing with a night time nap may not help enough to keep fatigue risk low and performance adequate.

There is also one more nuance to consider with biphasic sleep.  Fatigue levels, and possibly performance, may not be consistently at acceptable levels across the whole waking period.  Shift-workers with biphasic sleep can be sleepier during the first 1/2 of the waking period than during the 2nd half of the waking period when compared to those who sleep in one consolidated period[2].  If one portion of your split shift occurs when your shift-workers are sleepy, you may have an increased risk to manage.

But there is a bigger problem.  Shift-workers don’t like split shifts and they find ways to avoid the shifts they don’t like.  They will call in sick[3] which forces managers to scramble to fill the void or the shift-workers will swap shifts.  Although swapping shifts doesn’t leave a void, it can mess things up for managers who design tasks and work patterns with specific shift-workers and skill sets in mind, or with fatigue risk mitigating scheduling practices built in to the shift pattern.  In these cases, the work quality or efficiency can suffer or unanticipated fatigue risk can surface.  

From a Fatigue Manager’s perspective, my advice is that if you need to split shifts, only do it for day shifts and if you are going to do it, make sure that your shift-workers can find some personal advantages to splitting shifts otherwise they will “titrate” their exposure to the splits and mess up your well planned schedules.   For me, counting on shift-workers to nap during the day with our current organizational culture that does not really prioritize sleep and proactive fatigue management, it is too risky. I avoid split shifts that might break up our shift-workers’ night sleep at all costs.

One other caveat for condoning split sleep is that it can lead people to believe that splitting sleep into numerous periods can also be fine. This is called polyphasic sleep and it is a “no-no”.  The National Sleep Foundation published a consensus paper recommending against its use because of the negative physical and mental health consequences and performance outcomes[4].

Don`t copy text!