Choosing the “Right” Fatigue Monitoring and Detection Technology

Fatigue can shorten concentration, slow reaction times and impair decision-making skills resulting in increased health and safety risks for workers. It has been estimated that one in five fatal motor vehicle crashes in the U.S. can be attributed to fatigue. In addition, workers with sleep problems are 62% more likely to experience a work-related injury^1,2. While the adverse effects of fatigue are well known, the sources of fatigue can be attributed to a number of different factors, making it challenging to detect and manage³. To address this problem, there has been a rapid increase in fatigue detection devices currently available on the commercial market. With all of these newly available choices, it can be difficult to decide which technology would work best for you or your organization.

Importantly, fatigue monitoring and detection technologies (FMDT) can be broadly categorized into two types: 1) technologies that predict future fatigue based on the users’ recent sleep, work hours, etc. and 2) those that monitor and detect potential current fatigue using biological measures (e.g. eyelid movement and blink rate, reaction times to tests) or performance measures (e.g. lane departures). Detection technologies have become increasingly common in most new, higher end motor vehicles to alert drivers when they appear to be driving drowsy. However, these technologies are not limited to transportation. Some are available as wearable devices or mobile apps that can be used in almost any situation.

In choosing among the available fatigue detection technologies, there are several considerations, aside from cost. Here are seven of the most important:

Objective – what is the purpose for implementing fatigue detection? Determining the primary goal and intended outcome of the initiative is critical and will guide the selection of the right technology. Some questions to consider include: What data do you already have to help you understand the hazard? How will the FMDT technology, or the combined use of multiple FMDT technologies, contribute to achieving the goal? What do you plan to do with the additional data collected by the FMDT?

Validity – has the device been scientifically tested against other known fatigue measures to ensure it measures what it claims to measure? You can ask the supplier to provide results of scientific testing of their device. If the supplier doesn’t have validation information available, or is dismissive of the need for this research, you may want to broaden your search.

Reliability and generalizability – does the device work reliably and consistently in conditions where it will be used? For example, will it be able to withstand extreme temperatures or moisture if outdoor work is required? For eye-monitoring devices, will it be able to work as well under low light conditions, as it does in full daylight?

Sensitivity and specificity – does the device correctly identify “true” and “false” fatigue readings? Are these readings meaningful and beneficial? If a device does not provide accurate readings, users may lose trust in it, making the device not effective. Similarly, too many alerts may lead to the warnings being ignored.

Compatibility – is the device compatible with existing systems? For example, some devices are able to tie into existing in-vehicle telematics to give a multi-layered approach to detecting fatigue. Some devices also provide a hybrid approach, combining the ability to predict and detect increased risk for fatigue. These can provide a broader range of protection than single purpose models.

Predictability and feedback – is there a time lag between the recognition of fatigue and the likelihood of a safety critical event? In some situations, such as with driving, immediate feedback is needed to prevent a crash or close call. Does the system use multiple or appropriate cues (e.g., visual, audio, vibration) to alert the user? This can be especially helpful in situations where the worker may not be regularly looking at the device or in noisy situations. Does it send an alert to a safety manager to step in when a high level of fatigue is detected for a worker?

User acceptability – probably one of the most important considerations is if users will accept the technology into their regular workplace activities. If the device is uncomfortable or is cumbersome, it may not be used. Workers may also be concerned about their privacy, as this data is collected specific to each individual. Employers will need to keep this data secure and work with employees to build a system of trust. This can include assuring that the data will only be used for safety, and not disciplinary, purposes.

Choosing an effective fatigue detection device is unique to each workplace and can be a large financial investment. While the choices can be overwhelming, taking the time to research different technologies, understand their differences and tradeoffs, and working with management and labor representatives can ensure the “right” fit. Pilot testing the technology in a subgroup of the workforce prior to widespread adoption allows for time to become familiar with the device, examine its effectiveness and limitations at a small scale, and gain insight into any user acceptance issues. Most importantly, while these devices may be helpful, they can obscure efforts to address the underlying causes of fatigue and should not be used as the primary safety measure to reduce fatigue. Instead, they should be used judiciously and as a part of a more holistic fatigue management plan.

This blog is a part of a series from NIOSH’s Center for Work and Fatigue Research, Center for Motor Vehicle Safety and Center for Direct Reading and Sensor Technologies. Look for our next NIOSH Science Blog for tips for implementing a FMDT!

What other considerations for selecting fatigue detection technologies do you think are important? Are these technologies beneficial to workers and/or management, or are there any downsides? Have you successfully implemented FMDT in your fatigue management plan?

Imelda Wong, PhD, is the Coordinator for the NIOSH Center for Work and Fatigue Research

Kyla Retzer, MPH, is the Coordinator for the NIOSH Center for Motor Vehicle Safety

Emanuele Cauda, PhD, is the co-Director for the NIOSH Center for Direct Reading and Sensor Technologies

See the other blogs in this series.

The Who, What, How and When of Implementing Fatigue Monitoring and Detection Technologies

Setting Objectives for Fatigue Detection Technologies

We appreciate your input. Please do not mention company or product names in your comments. NIOSH cannot endorse or appear to endorse any product or company. As such, company or product names will be removed from comments prior to posting.

References

1. Tefft B. Prevalence of motor vehicle crashes involving drowsy drivers, United States, 2009-2013. 2014; https://aaafoundation.org/wp-content/uploads/2017/12/PrevalenceofMVCDrowsyDriversReport.pdf. Accessed November 2020.
2. Uehli K, Mehta A, Miedinger D, et al. Sleep problems and work injuries: a systematic review and meta-analysis. Sleep Med Rev. 2014;18(1):61-73.
3. Di Milia L, Smolensky MH, Costa G, Howarth HD, Ohayon MM, Philip P. Demographic factors, fatigue, and driving accidents: An examination of the published literature. Accident Analysis & Prevention. 2011;43(2):516-532.