Dossier - An Overview of RCAF Fatigue Issues and Selected Countermeasures

Image Gallery

Magazine Article / July 15, 2013 / Project number: 2013 - Issue 2

Co-Authors: Colonel Colin Keiver, Director Air Contracted Force Generation, Ottawa Lieutenant-Colonel Jason Stark, Commanding Officer 429 Squadron, 8 Wing Trenton

During the recent air campaign over Libya (Ops Aleta and Libeccio), our air transport aircrews supporting that campaign experienced some very serious fatigue-related flight safety incidents (FSI) which could easily have been accidents and were secondary to poor mission scheduling. In the case of a J-model CC130, they reached “top of climb” in a very hypoxic state, because cabin pressurization activation was overlooked in the checklist. In the case of a CC177 they had 3 FSIs in one mission: on day 2, hot brakes from a checklist oversight (with modelled fatigue levels resulting in cognitive effectiveness of 68.5%, in excess of equivalence to intoxication with alcohol to blood alcohol content (BAC) of 0.08%); on day 7, mis-identified an airfield during approach to landing (cognitive effectiveness of 45.3%, which is off the scale for BAC equivalence in the performance band where no one can function well on any task) and on day 10, a flap over-speed on landing (cognitive effectiveness of 68% (again equivalent to BAC in excess of 0.08%). Recently, a USAF C-17 was so fatigued they actually landed at the wrong airfield in spite of the runway being only 3,400 feet long instead of the 11,500 foot runway four miles to the south. Apparently, USAF crews are also vulnerable to flying poor mission schedules.

Generally, crew duty/crew rest regulations are based on operational experience but not on the foundational science that addresses the interaction of fatigue and circadian processes and theirs effects on performance1. The current rules are an attempt to give crews time to recover between flights and often have provisions to compensate crews for long crew days. However, they are not normally prescriptive for behaviour prior to missions, and they treat time of day equally throughout the 24-hour clock.

More recently, in most jurisdictions, there is recognition that aircrew performance during normal daylight hours is usually quite good. However, there is an evolving awareness that during night ops, people are working when melatonin produced by the body is released into their blood which facilitates sleep and induces drowsiness, thus impairing night performance, unless they are adapted to night operations.

For example, the latest FAA policy recognizes that during the WOCL (Window Of Circadian Low – about midnight to 0600 h) performance can be significantly impacted relative to daytime performance. Recently, there is also recognition that there is a complex interaction between “time of day” and “sleep opportunities” and “quality of the sleep environment” and “performance”. However, new modeling software (such as FASTTM) takes the mystery out of the equation since it recognizes the complex interactions between “work”, “sleep” and “time of day”.

Before illustrating the FASTTM models for the 2 RCAF missions referred to above, the following brief overview of the FASTTM modeling program is provided (see figure 1).

  • The vertical axis on the left side of the FASTTM graphs represents human cognitive performance effectiveness as a percentage of optimal performance (100%). The oscillating line in the diagram represents average performance (cognitive effectiveness) as determined by time of day, biological rhythms, time spent awake, and amount of sleep. This line is thin and black during periods spent awake, thin and gray during sleep periods and is a bold black line during work periods.
  • The dotted line, which is below the cognitive effectiveness curve and follows a similar oscillating pattern as the cognitive effectiveness curve, represents the 10th percentile confidence interval.
  • The green band (from 90% to 100%) represents acceptable cognitive performance effectiveness for workers conducting safety sensitive jobs (flying, driving, weapons operation, command and control, etc.).
  • The yellow performance band (from 65% to 90% cognitive effectiveness) indicates caution. Personnel engaged in skilled performance activities such as aviation should not be allowed to operate within this performance band.
  • The pink performance band (below 65%) represents seriously impaired performance effectiveness, for example what might be expected after 2 days and a night of sleep deprivation. Under these conditions, no one can be expected to function well on any task.
  • A value of 77% cognitive effectiveness corresponds to performance with a (BAC) of 0.05% (legally impaired in some jurisdictions). A value of 70% cognitive effectiveness corresponds to a BAC of 0.08% (legally impaired in most jurisdictions). These BAC equivalency levels associated with sleep deprivation/fatigue are based on three important studies2-4.
  • The abscissa (x-axis) illustrates periods of work (red bars), sleep (blue bars), darkness (gray bars) and time of day in hours.
  • The grey triangles labelled located just above the abscissa are event markers indicating the key waypoints of this mission (including latitude and longitude to reflect the photoperiod (read sunrise and sundown times) to reconcile circadian stresses as a function of changing time zones.
  • Red triangles flag flight safety incidents.

Figure 1 model illustrates the mission schedule flown by CC177 crew and reflects the worst aircrew cognitive effectiveness levels we have seen over the 10 years we have been modeling RCAF Air Operations. It is hardly surprising that there are 3 flight safety incidents flagged in this model (see red triangles, C1, C2, and C3 which correspond to hot brakes from a checklist oversight, misidentified airfield on final approach to landing, and flap over-speed on landing respectively). These dangerous levels of aircrew performance can be, and must be, avoided by better flight scheduling and better aircrew sleep hygiene.

Narrative from the Aircraft Commander

On arrival to OLBA April 3rd, had hot brakes due to an oversight in the approach briefing phase. This could be fatigue related as I have about 1000 hours on the airframe and 2400 total military hours with an additional 1200 civilian hours.

Next event occurred on arrival to ETAD on the 8th April where I misidentified an airfield (EDFH Frankfurt Hahn) that was not our intended arrival airport. Distance between the 2 airports is 30 miles; it was a clear VFR day. Continued to follow ATC vector and descent instruction but due to the misidentification I descended to the next altitude very quickly so that I could carry out an approach to that misidentified airfield. End result was being low and loud due to aircraft configuration, and after some embarrassment, landing at the proper airport. Next event occurred on arrival to OAKN on 11th April, 10-15 knot flap over-speed on selection of 3/4 flaps for landing. Remainder of mission was uneventful.

Figure 2 illustrates a mission flown by a CC130J crew. This illustrates modeled performance over 11 separate flights. Except for the first two flights (Trenton to Bagotville and Bagotville to Halifax), the op tempo of this mission produced very deleterious levels of cognitive effectiveness in that about 82% of the time of all 11 flights were spent at performance levels equivalent to being very intoxicated with blood alcohol levels off the scale (see right hand vertical axis). During the last 6 flights, modeled performance was especially worrisome and ranged from 48% to 60%. The red triangle flags the flight safety incident on March 27th (approximately 1712 hrs zulu) where a checklist item was overlooked thus producing a hypoxic cabin altitude.

This model reflects the 2nd worst performance we have modeled from CF air transport operations. These levels of performance can easily result in accidents, and can be avoided by better scheduling.

Narrative by Aircraft Commander

Crew entered 4-hr standby at 0600L (1000z) on 21 Mar 11. Crew was alerted to depart at 1700L (2100z) on 21 Mar 11. We departed CFB Trenton at 2100L (2359z) on 21 Mar with the intention of arriving at Prestwick, Scotland at 1100L (1100z) the following day. However, due to snow showers at CFB Bagotville, the crew was unable to depart and entered crew rest at 0100L (0400z) on 22 Mar.

8 Wing Ops was called and informed that the crew would be ready to depart at 1500L (1800z) after getting 12 hours crew rest. The crew was told to anticipate that departure and timed their rest in order to be ready. Around 1000L (1300z), I received emails and phone calls regarding changes in mission that required my immediate attention and technically interrupting my crew rest. I did have 8 hours uninterrupted but the work started my crew day at 1100L (1400z). I received direction from three sources: 8 Wing Ops, AOC Winnipeg, and 436 Sqn Ops. Our takeoff was delayed until 1859L (2159z) which was already 6 hours into the crew day. Due to severe turbulence throughout the Atlantic, the crew had to divert to Halifax in order to fly south of the area of turbulence as we crossed the Atlantic for Prestwick. We landed 0848L (0848z), and took over an hour to secure the aircraft and get to the hotel. Our crew duty day was 18 hours.

We entered crew rest at 1000L (1000z) and slept throughout the day. We were told to expect a 24 Mar, 0100L (0100z) departure to return to CFB Trenton. Around 2100L (2100z), I received emails and phone calls changing the mission to a 0730L (0730z) departure on 24 Mar. My crew had rested for the early morning departure and therefore found it difficult to properly rest for the 6.5 hour change in itinerary. Due to our sleep/rest cycle, none of my crew was able to sleep through the night and we were quite tired. We started our day at 0430L (0430z). We departed for CFB Trenton, arrived in Goose Bay, and diverted to CFB Greenwood to pick up another load destined for NAS Sigonella. We arrived at CFB Greenwood at 1401L (1701z), but continued to work and load the aircraft in preparation for the next day. We arrived at the hotel at 1700L (2000z) and entered crew rest. The total day was 15.5 hours. However, if you account for our planned itinerary departure, which we were crew-rested for, it was a 21 hour day. At this point, I began to see signs of exhaustion in our crew, but since we didn’t technically have two days >16 hours, I did not ask for 36 hours of rest. We had 24 hours to rest at CFB Greenwood, however, our sleep/rest cycle was still set to East Coast time so the crew was awake and active at 1000L (1300z).

We began 25 Mar at 1700L (2000z) to depart CFB Greenwood. Again, I received phone calls and emails that dealt with mission changes while in crew rest. Also, a fourth controlling agency was introduced (the ALCE in Prestwick, Scotland) which led to more confusion about who was controlling our mission/itinerary. After arriving at the airfield, we found out that we were to download the equipment on the aircraft that was loaded the previous day and divert into CFB Bagotville prior to departing for RAF Kinloss. We made an on-time departure at 1950L (2250z) and arrived at CFB Bagotville. There were snow showers there but we got a break in the snow, de-iced and departed at 2331L (0231z). We landed in RAF Kinloss at 1136L (1136z), but due to lack of availability of MAMS personnel, we had to load the aircraft. We arrived at the hotel and entered crew rest at 1400L (1400z). Our crew day was 18 hours, however, due to our sleep/rest cycle the crew had been awake and active for ~25 hours.

We began 27 Mar at 0700L (0600z) after 16 hours of crew rest. Again, I dealt with itinerary changes while in crew rest. We departed RAF Kinloss for NAS Sigonella on our fourth long itinerary day in a row. In my opinion, this was a contributing factor in our flight safety incident that occurred between NAS Sigonella and Trapani AB. During this leg, the co-pilot skipped a checklist step (turning on the Flight Deck and Cargo Compartment AC) which led to a cabin altitude of >10,000 ft. Passing 10,000 ft, I checked that the aircraft was pressurizing and noted that the cabin was pressurizing but the cabin altitude was higher than usual and decided to check it again at level off. Before we leveled, we received the Cabin Altitude High ACAWS. After going on oxygen, an emergency descent, and a bit of trouble shooting, the co-pilot error was discovered. The aircraft pressurized normally and we continued the mission. After landing at Trapani AB, I reported the Flight Safety. My co-pilot continued to make checklist type errors and stated that he was tired throughout the day. While on the ground in Trapani, our mission changed 4 times and we were finally told to depart for Prestwick, Scotland. We told 8 Wing Ops that we couldn’t make Prestwick due to duty day (planned landing at 19+ hours of crew day) and that we would prefer to stay at Trapani. We were told that there were no rooms available in Trapani and we needed to depart. We decided on Frankfurt-Hahn (it was previously our destination and had coordinated parking/rooms). We arrived at 0158L (2358z) and entered crew rest at 0330L (0130z). Our crew day was 19.5 hours.

On our flight into Frankfurt-Hahn, we had a long discussion about crew rest requirements and how we were feeling. The crew was very tired at this point. We decided to take a short day the next day and changed our itinerary to fly to Prestwick, Scotland. We began 28 Mar at 1530L (1330z) and departed at 1648L (1448z). We landed at 1806L (1706z) and entered crew rest at 2000L (1900z).

We began 29 Mar at 0900L (0800z). Again, I received phone calls/emails during crew rest. Originally we were to depart at 1100L (1000z) but that was changed to 1300L (1200z). We were forced to stop in Goose Bay to get fuel (due to winds) and arrived at CFB Trenton at 1809 L (2209z). We departed the base and ended our day at 1930L (2330z). Our final crew day was 15.5 hours.

My biggest issue was the constant interruption of crew rest. The controlling agencies disregarded the crews need for 8 hours of uninterrupted sleep. This was a contributing factor for the Flight Safety incident that occurred. Second was the constant change in our sleep/rest cycle. On two of the days, the crew was awake and active for over 24 hours. Third was that we never understood who was controlling the mission. We had 4 agencies telling us four different pieces of information.

Overall, I feel that the mission was safely executed. This way the controlling agency knows when a crew should be unreachable. Secondly, if there is any chance that an itinerary could change while crews are resting (which was the norm in this case), then they should be put on Standby status vice being told to be ready for a set departure time. Crews handle standby status differently and can manage their sleep/rest cycle to be ready to execute a mission at any time. If a crew is told of a departure time, then rest is managed so that they can maximize it for that particular departure time. Any slip in departure will result in awake/active times of greater than 24 hours.

The very best aircrew fatigue countermeasure is optimal mission scheduling that will not unnecessarily compromise aircrew performance. To facilitate optimal mission scheduling, we have recommended that the RCAF acquire FASTTM software. The idea is to merge FASTTM with the recently acquired new scheduling software (Airlift Planning Tool or APT), where FASTTM would communicate in background with APT. The scheduler would develop mission schedules normally and then hit a ‘hot key’ to bring up the schedule in FASTTM automatically. This would allow the scheduler to identify times in the schedule when modeled performance would be below acceptable levels. This would provide an opportunity to optimize the schedule (and thus limit unnecessary aircrew fatigue) before a squadron is tasked to execute the mission. When foreign policy imperatives that drive military taskings dictate high priority immediate response, and there is no room to optimize and/or delay the mission, FASTTM software would identify specific times in the itinerary when performance would be impaired, which is preferable to being unaware of the problem, or worse, ignoring it. In this case, the Flight Surgeon community can use countermeasures to facilitate aircrew sleep, for example with melatonin or zopiclone, during times when sleep has to occur during physiologic day or sustain alertness e.g., with caffeinated gum – StayAlertTM. At present, other alertness medications such as modafinil or dexaphetamine are not approved for use in RCAF operations. Similarly, optimum shift schedules and the pharmaceutical fatigue countermeasures can be used to support aircraft maintainers, and Air Operations Centre personnel.

In response to significant operational fatigue problems over the last 15 years (mainly Bosnia and Afghanistan), the RCAF has invested heavily in fatigue research 1, 5-14. The most recent fatigue countermeasure project was focused on circadian phase shifting.

Circadian Phase Shifting

This involves the manipulation of circadian rhythms, either forwards or backwards, to counter jetlag and shiftlag. The two modalities employed to phase shift are 1) appropriately timed ingestion of melatonin to overlap with the body’s production of melatonin and 2) appropriately timed light treatment during physiologic night (i.e., when the body is producing melatonin) since light treatment will transform physiologic night into physiologic day by suppressing the body’s production of melatonin. Another important factor is avoidance of light at a time when exposure to light will be counterproductive to the desired phase shift. Part of the Air Force’s investment in fatigue research was to fund a comprehensive project to optimize our ability to shift circadian rhythms. This project has been very successful. Three of our four circadian publications were awarded three international awards for “Best of Sleep Medicine” for each of 2009, 2010, and 2011 12,13,15. The main output from this work was to optimize Phase Response Curves (PRCs) for each of light and melatonin. These PRCs can be used to generate phase shifting treatments, 2 of which are illustrated below.

The phase shifting treatment in Figure 3 Model above is to phase advance aircrew from Trenton to Camp Mirage. The yellow horizontal bands represent the photoperiod (read daylight hours) in Trenton and the 2 destinations en route to Camp Mirage. The red vertical bar within each yellow band represents sunrise and sunset in each of the 3 locations, thus illustrating when sunlight can be accessed and when a light treatment device has to be used when the sun is not up.

This phase advance treatment involves a 0.5 mg dose of melatonin taken about 2 hours before the onset of the body’s melatonin rhythm. Thus, on the first night of treatment the melatonin dose is taken at 1900 hours for someone who has a normal bedtime of 2300 to 2400 with a 7 to 8 hour time in bed. Upon arising the individual undergoing this treatment would seek out bright light for 3 hours. This would yield about an hour of phase advance for each day of treatment. Thus, to remain on the “sweet spots” of the PRCs for light and melatonin, treatment times and bedtimes will advance by 1 hour for each subject’s day of treatment. The black triangle is the time at which the body temperature reaches its daily minimum and is a marker for keeping track of the direction and magnitude of the phase shift from day to day. An individual following this treatment would arrive in Camp Mirage about an hour out of synch with the local photoperiod. He/she would need a day to recover from the flight and would continue treatment for the first day in Camp Mirage thus being able to report to Ops the following day completely on local time with no jetlag. 

By going to bed earlier and earlier each successive day, while at home in the last 5 days before deployment. For those who want more time with their families immediately before deploying, an alternative approach is to shift the long way around the clock (i.e., phase delay of 15 hours versus phase advance by 9 hours) as shown in Figure 4. Since the body is normally 50% more effective at phase delay than phase advance the phase delay option for Camp Mirage would only take an additional day of treatment in Camp Mirage before being free of any residual jetlag.

The phase delay treatment figure 4 involves light at night and melatonin in the morning upon awakening, where the individual being treated would remain in dim light (D=dark) after the morning melatonin dose. In the above treatment schedule, the small black arrows represent ideal timings for naps to more easily stay awake during the nightly light treatment.

Circadian desynchrony that is inherent in rapid deployment across multiple time zones or in shiftlag. To counter jetlag and shiftlag, Flight Surgeons can recommend appropriately-timed ingestion of specific melatonin formulations, and appropriately-timed light treatment, as well as avoidance of light at certain times. Since generating the treatment grids above is labour and time intensive, in the near future, we are expecting to develop an application that will allow individuals to input their departure and destination locations and travel dates to receive comprehensive phase shifting treatments for either phase advance of phase delay.

Recent operations have demonstrated that the RCAF has not mastered the art of Air Mobility across time zones mission scheduling. Crews are being exposed to unsafe situations that could be mitigated through scheduling and science. There will always be the missions that require the 150% effort and in these situations smart scheduling and FAST will clearly identify the flight safety risk assumed by HHQ. Authorizing the mission would then place the assumption of risk at the level of Commanders vice Aircraft Commanders. In this era of reduced crew experience and fiscal constraints, it is imperative that we work smart and efficiently with the precious resources we have.

Currently, the Air Force is looking into modifying scheduling by utilizing a fatigue-modelling tool such as FAST in conjunction with operational scheduling software. The RCAF aeromedical community has promulgated a Flight Surgeon Guideline 16 to provide guidance to Flight Surgeons in managing fatigue from a medical perspective, including screening for underlying medical conditions which contribute to fatigue, and prescribing countermeasures to assist with sleep, alertness and circadian phase shifting.

References:

  1. J.A., C. and Caldwell, J. L. (2003). Fatigue in aviation: A guide to staying awake. Ashgate limited publishing.
  2. Arnedt, J. T., Wilde, G. J. S., Munt, P. W., and MacLean, A. W. (2001). How do prolonged wakefulness and alcohol compare in the decrements they produce on a simulated driving task? Accident Analylsis and Prevention, 33, 337-344.
  3. Dawson, D. and Reid, K. (1997). Fatigue, alcohol and performance impairment. Nature, 388, 235.
  4. Lamond, N. and Dawson, D. (1999). Quantifying the performance impairment associated with fatigue. Journal of Sleep Research, 8 (no. 4), 255-262.
  5. Paul, M. A., Brown, G., Buguet, A., Gray, G., Pigeau, R. A., Weinberg, H., and Radomski, M. (2001). Melatonin and zopiclone as pharmacologic aids to facilitate crew rest. Aviat Space Environ Med, 72 (11), 974-984.
  6. Paul, M. A., Gray, G., Kenny, G., and Pigeau, R. (2003). Impact of melatonin, zaleplon, zopiclone, and temazepam on psychomotor performance. Aviat Space Environ Med, 74 (12), 1263-1270.
  7. Paul, M. A., Gray, G., MacLellan, M., and Pigeau, R. A. (2004). Sleep-inducing pharmaceuticals: A comparison of melatonin, zaleplon, zopiclone, and temazepam. Aviat Space Environ Med, 75 (6), 512-19.
  8. Paul, M. A., Gray, G. W., Lieberman, H. R., Love, R. J., Miller, J. C., and Arendt, J. (2010). Management of circadian desynchrony (jetlag and shiftlag) in cf air operations. TR 2010-002, DRDC Toronto.
  9. Paul, M. A., Gray, G. W., Sardana, T. M., and Pigeau, R. A. (2003). Fatigue countermeasures in support of cf cc130 air transport operations: From the operation to the laboratory and back to the operation. (TR 2003-106). Defence Research and Development Canada--Toronto, 2003; Report No. TR 2003-106.
  10. Paul, M. A. and Miller, J. C. (2004). Fatigue assessment in camp mirage cc130 aircrew: Recommendations for pharmacologic intervention. (TR 2004-021). TR 2004-021. DRDC Toronto.
  11. Paul, M. A., Miller, J. C., Gray, G. W., Buick, F., Blazeski, S., and Arendt, J. (2007). Phototherapy for circadian phase delay: A comparison of 4 phototherapeutic devices. Aviat Space Environ Med, 78 (under review), 645-652.
  12. Paul, M. A., Miller, J. C., Gray, G. W., Love, R. J., Lieberman, H. R., and Arendt, J. (2010). Melatonin treatment for eastward and westward travel preparation. Psychopharmacology, 208 (3), 377-387.
  13. Paul, M. A., Miller, J. C., Love, R. J., Lieberman, H. R., Blazeski, S., and Arendt, J. (2009). Timing light treatment for eastward and westward travel preparation. Chronobiology International, 26 (5), 867-890.
  14. Paul, M. A., Pigeau, R. A., and Weinberg, H. (1998). Human factors of cc-130 operations. Volume 6: Fatigue in long-haul re-supply missions. Toronto, canada. (Report No. 98-R-19). Defence and Civil Institute of Environmental Medicine; DCIEM Report No. 98-R-19.
  15. Paul, M. A., Gray, G. W., Lieberman, H. R., Love, R. J., Miller, J. C., Trouborst, M., and Arendt, J. (2011). Phase advance with separate and combined melatonin and light treatment. Psychopharmacology, DOI: 10.1007/s00213-010-2059-5.
  16. Flight Surgeon Guideline 1400-03 Fatigue Management in Aircrew. http://winnipeg.mil.ca/cms/Libraries/Flight_Surgeon_Guidelines/FSG_1400-03_FATIGUE_ MANAGEMENT_IN_AIRCREW.sflb.ashx
Date modified: