translating fatigue research into technologic countermeasures · translating fatigue research into...

Translating Fatigue Research into Technologic Countermeasures

David A. Lombardi, PhD

Principal Research Scientist Center for Injury Epidemiology,

Liberty Mutual Research Institute for Safety

Co-Director, Occupational Injury Prevention Research Training Program(OIPRT)

Harvard Education and Research Center (ERC)

Presented at “Sleep and Shift Work: Optimizing Productivity and Health Management in the 24/7 Global Economy”, Harvard School of Public Health, September 28, 2012

Presentation for Review : 09/12/201

Learning Objectives

1. Participants will be able to demonstrate an understanding of a conceptual model of fatigue and implications on safety

2. Participants will be able to understand approaches to mitigate workplace fatigue, including basic attributes, challenges and opportunities of technical countermeasures

3. Participants will understand the various measures of fatigue that include physiological, behavioral, subjective self-report and performance measures

4. Participants will be able to explain Haddon’s Matrix and how it relates to technological countermeasures


“Fatigue is a biological drive for recuperative rest” (Safe recovery)

Fatigue

Time of day

Time/s awake

Task-related factors

Impaired Performance Capabilities

“Accident”

Rest

Sleep

Sleepiness

Williamson A, Lombardi DA, et al., (2011). The link between fatigue and safety. Accid. Anal. Prev. 43:498-515.

• Slowed reaction times • Lapses of attention • Errors of omission • Compromised problem

solving ability

•Sleep homeostasis


Trends in Sleep Duration in the US

• Percentage of Adults Aged >18 Years Who Reported an Average of <6 Hours of Sleep per 24-Hour Period, by Sex and Age Group --- National Health Interview Survey, United States, 1985 and 2006

Source: MMWR, February 29, 2008 / 57(08);209.


Self-perception of Sleepiness

Source: Drake CL, Roehrs TA, Richardson GS, Roth T. Epidemiology and morbidity of excessive daytime sleepiness. Sleep. 2002; A124.I.

Measurement Subjective-Epworth Sleepiness Scale Objective-Multiple Sleep Latency Test


Preventing and Managing Fatigue

• Can operator (e.g., driver, pilot, conductor etc.)

fatigue be prevented or effectively managed to significantly reduce "accident" or performance error risk?


Approaches to Mitigate Workplace Fatigue • Industry standards and best practices (e.g. ACGME)

• Regulatory (Gander et al., 2011)

– Hours of service regulations » FAA: regulations for domestic flights generally limit pilots to eight

hours of flight time during a 24-hour period

» FMCSA: CMV 11-Hour Driving Limit, May drive a maximum of 11 hours after 10 consecutive hours off duty...

• Organizational (Folkard and Lombardi, 2006)

– Designing safer work and “break” schedules (e.g. “risk index” models)

• Technological (Balkin et al., 2011)

– Approaches in which “objectively detect or predict operator fatigue” used to complement or supplant organizational or regulatory approaches

Low Tech

High Tech


The State of Fatigue Interventions

• “The scientific understanding of fatigue, sleep, shift work, and circadian physiology has advanced significantly over the past several decades, current regulations and industry practices have in large part failed to adequately incorporate the new knowledge.” (Dinges, 1996, Caldwell et al., 2009)


Technological Countermeasures

• Frequently prescribed as countermeasures for fatigue in:

• Road transport – Truck drivers • Aviation – Pilots, Air traffic controllers,

security personnel • Maritime - Seafarers • Railway - Conductors • Aerospace- Astronauts • Plant operations – Nuclear plants


Ideally includes:

1. Ability to predict fatigue – based on the factors that produce it (i.e., take

into consideration sleep patterns and circadian rhythm)

– prior to any impact on operational performance (i.e., safety and productivity)

Source: Balkin TJ, Horrey WJ, Graeber RC, Czeisler CA, Dinges DF. The challenges and opportunities of technological approaches to fatigue management. Accid Anal Prev. 2011 Mar;43(2):565-72.

Technological Approaches (Balkin et al., 2011)


2. Can measure and monitor fatigue/performance online in the operational environment – as a backup and check of the performance prediction model

3. Can effectively intervene when potential deficits are identified or anticipated – with interventions calibrated to restore and sustain

alertness/performance as long as needed

» until operator can obtain adequate recuperative rest

Technological Approaches (Balkin et al., 2011)


Should be: • Valid • Reliable • Sensitive (detect fatigue) and specific (minimize false

negatives) • Generalizable • “Assessments of these criteria should be done in the actual

operational environment” – Most fatigue-detection technologies unproven in real world

situations (Brown, 1997, Balkin et al., 2011), with rare exceptions (Dinges et al., 2005)

Source: Dinges, D.F., Mallis, M.M., 1998. Managing fatigue by drowsiness detection: can technological promises be realized? In: Hartley, L.R. (Ed.), Managing Fatigue in Transportation: Proceedings of the Third International Conference on Fatigue and Transportation. Elsevier, Oxford

Ideal fatigue-detection system (Dinges and Mallis, 2008)


Adapted from Trochim, William M. The Research Methods Knowledge Base, 2nd Edition.

Review: Concepts in Measurement

“Measurement focuses on the crucial relationship between the empirically grounded indicator (i.e., the observable response) and the underlying unobservable concept” (Carmines & Zeller, 1979)


Hancock and Verwey (1997) propose that,:

• “Automated systems need to exhibit different types and levels of adaptation, so as to facilitate application to a wide sample of users” – should follow design guidelines, based on knowledge of

human capabilities and limitations (e.g., static adaptation)

– should take into account the effects of dynamic, external variables on the average operator (e.g. generalized dynamic adaptation)

– should take into account the capabilities and limitations of the current operator (e.g. idiosyncratic dynamic adaptation)

Source: Hancock, P.A., Verwey, W.B., 1997. Fatigue, workload and adaptive driver systems. Accident Analysis and Prevention 29 (4), 495–506.

Ideal fatigue-detection systems


Technological Countermeasures

Other challenges:

• Human–automation interaction requires user acceptance and compliance (Balkin et al., 2011)

• In real life?

• “…nearly half of the surveyed truck drivers expressed a negative view towards developing a technological countermeasure against driver fatigue. The negative view was not related to personal experiences of fatigue-related problems while driving.” (Häkkinen et al., 2001)


Types of fatigue measures: 1. Physiological Measures 2. Behavioral Measures 3. Self-Report Measures (subjective) 4. Performance Measures

“Primarily designed to be used in the laboratory and have varying levels of utility in the workplace.” (Sherry, 2000)

Source: Sherry P. (2000). Fatigue Countermeasures in the Railroad Industry: Past and Current Developments. Association of American Railroads, Washington, DC.

Overview: Measurements of Fatigue


Operator-centered approaches (Dinges et al., 1998, Balkin et al. 2001)

• Readiness-to-perform and fitness-for-duty technologies (e.g., PVT, OSPAT, Fit 2000, ART90, etc.)

Primary question: Does the operator have sufficient alertness prior to a work cycle?

– Performance-based: directly measures the cognitive processes and motor skills required for a specific safety-sensitive job

– Physiological–based: assess involuntary signs that stressors may produce due to their effect on the brain

» “ability to process afferent impulses and generate appropriate involuntary efferent output”

Overview of fatigue monitoring and detection approaches



• Online operator monitoring technologies

• Primary question: Does the operator have sufficient alertness during the work cycle?

– EEG based algorithms (measuring brain wave activity, e.g., increases in alpha, beta, delta and theta waves)

– Ocular (PERCLOS, MOMS, examines percent eye closure, measures of saccadic velocity, pupillometics, see Dinges 1998 and 2005)

– Actigraphy (reliable in estimating sleep/wake timing and duration for fatigue/prediction models, but not direct monitoring)

– Video based (currently used to identify fatigue as a factor in triggering and event post-hoc with low real-time predictive ability)




• Performance-based monitoring technologies – a proxy for fatigue that identify behaviors and performance that

may lead to unsafe conditions

» lane tracking performance

» reaction time changes

– Limitations, subject to false alarms

• Biomathematical models of fatigue/alertness – Primary question: Is the structure of the operators work schedule

maximized to be the safest possible, based upon known factors about the operator, environment, and tasks



• Useful in the prediction of performance or effectiveness levels when comparing different operational schedules (Mallis et al., 2004; Dinges et al., 2004)

• Used as support tools within Fatigue Risk Management Systems (FRMS)

– Two process model, Three-process models

– SAFE (System for Aircrew Fatigue Evaluation) model: developed for aviation operations

– Sleep, Activity and Fatigue Task Effectiveness (SAFTE) model: developed for military and industrial settings

– Fatigue Avoidance Scheduling Tool (FAST): designed to help optimize the operational management of aviation ground and flight crew

– Interactive Neurobehavioral Models

– Faid Model, CAS Model

– “Risk Index” Model

Brief overview of biomathematical models of alertness and fatigue


Criteria in Evaluating Technical Countermeasures (Barr et al., 2009)

• Scientific and Engineering Guidelines – Environmental: does the technology operate accurately

and reliably in real world conditions (lighting, temperature, humidity, vibration, etc.)?

– Reliability and Validity: does the technology have positive and negative predictive value, sensitivity and specificity?

– Anthropometric: can the technology be generalized across a broad range of users?

– Engineering Design: how much maintenance does the technology require?


Criteria in Evaluating Technical Countermeasures (Barr et al., 2009)

• User Acceptance Elements – Ease of Use: does the technology operate in real time, is it

invasive, does it accommodate eyewear, provide a sufficient alert?

– Ease of Learning: is training minimal allowing for adequate and reliable memory of proper use?

– Perceived Value: does the user trust the feedback of the technology, providing a sense of efficacy with limited risk of usage?

– Advocacy: will device be adopted, recommended and purchased by a sufficient number of users?

– User Behavior: is the user distracted by the technology and do they adapt to using it effectively?


Haddon Matrix Applied to Road Safety

Haddon Jr W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Report, 1980, 95:411–421.

Haddon matrix implies that injury producing events occur over time and are modifiable.


Technological Countermeasures Case Study

• TBD


References

Barr L, Popkin S, Howarth H, Carroll RJ . (2009). An Evaluation of Emerging Driver Fatigue Detection Measures and Technologies: Final Report, FMCSA-RRR-09-005 (Washington, DC: U.S. Department of Transportation, Federal Motor Carrier Safety Administration).

Brown ID. (1994). Driver Fatigue. Human Factors, 36 (2), 298-314.

Balkin TJ, Horrey WJ, Graeber RC, Czeisler CA, Dinges DF. The challenges and opportunities of technological approaches to fatigue management. Accid Anal Prev. 2011 Mar;43(2):565-72.

Dinges DF. (2008). “Fatigue: Where Biology Meets Technology“, Keynote. FAA Fatigue Symposium, Tysons Corner, VA.

Dinges, D.F., Mallis, M., Maislin, G., Powell, J.W., 1998. Final Report: Evaluation of Techniques for Ocular Measurement as an Index of Fatigue and as the Basis for Alertness Management (Report No. DOT HS 808 762). National Highway Traffic Safety Administration, Washington, DC.

Dinges, D.F., Rider, R.L., Dorrian, J., McGlinchey, E.L., Rogers, N.L., Cizman, Z., et al., 2005b. Optical computer recognition of facial expressions associated with stress induced by performance demands. Aviation, Space and Environmental Medicine, 76 (6), B172–182.

Folkard S, Lombardi DA. Towards a “Risk Index” to assess work schedules. Chronobiology International: The Journal of Biological & Medical Rhythm Research. 2004; 21(6): 1063 – 1072.


References (continued)

Folkard, S, Lombardi, DA, “Modeling the Impact of the Components of Long Work Hours on Injuries and ‘Accidents’,” American Journal of Industrial Medicine, Vol. 49, No. 11, pp. 953-963, 2006

Folkard, S, Lombardi, DA, and Spencer, M., “Estimating the Circadian Rhythm in the Risk o f Occupational Injuries and ‘Accidents’,” Chronobiology International, Vol. 23, No. 6, pp. 1181-1192, 2006

Haddon Jr W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Report, 1980, 95:411–421.

Häkkinen H, Summala H. Fatal traffic accidents among trailer truck drivers and accident causes as viewed by other truck drivers. Acid Anal Prev. 2001 Mar;33(2):187-96.

Lombardi DA, Folkard S, Willets JL. Smith GS. Daily sleep, weekly working hours, and risk of work-related injury: US National Health Interview Survey (2004-2008). Chronobiol Int, 2010 Jul; 27(5):1013-30.

May JF, Baldwin CL. (2009). Driver fatigue: The importance of identifying causal factors of fatigue when considering detection and countermeasure technologies. Transportation Research Part F: Traffic Psychology and Behaviour, 12(3), 218-224.

Sherry P. (2000). Fatigue Countermeasures in the Railroad Industry: Past and Current Developments. Association of American Railroads, Washington, DC.


Questions or Comments?

David A. Lombardi, Ph.D. Principal Research Scientist Center for Injury Epidemiology email: [email protected] Liberty Mutual Research Institute for Safety Center for Injury Epidemiology 71 Frankland Road Hopkinton, MA 01748 phone: (508) 497-0210 fax: (508) 435-3456


translating fatigue research into technologic countermeasures · translating fatigue research into...

Documents