Scientific Validity
We take pride in our scientifically validated fatigue risk management solutions. Whether it’s the Federal Railway Administration, academic journals, or operational tests, we encourage the publication of testing of our models and technology to assure you we offer the best solutions in the world.
Validation of Wrist Actigraphy
The gold standard for assessing sleep in research and hospital settings is polysomnography. In real world settings, where polysomnography is impractical, actigraphy provides an accurate estimate of sleep patterns in normal, healthy adult populations and patients suspected of certain sleep disorders. The ReadiBandTM relies on actigraphy to accurately track the primary fatigue factors so that this data can be fed into the SAFTE® model to assess fatigue risk.
Validation of Fatigue Science Readiband Technology
Fatigue Science is committed to scientific validation of our technology. Click here to view a presentation of the ReadiBand system and the related science. Click here
Validation of the Safte Model
The SAFTE® model was chosen as the best fatigue-risk prediction tool because it is a physiological approximation that incorporates a number of improvements compared to the prior models. In general, those changes were designed to improve conformance with the underlying principles that form the basis of performance effectiveness predictions. The model includes a realistic representation of the underlying circadian processes, a sophisticated routine governing the intensity of sleep as a function of time of day, and consideration of sleep inertia. To validate the model, the predictions of the model for the effects of total sleep deprivation were compared to an independent set of data reported by Angus and Heslegrave [2].
Figure 4: SAFTE™ Model predictions for cognitive performance under total sleep deprivation (solid line) compared to mean normalized cognitive performance (filled squares) reported by Angus & Heslegrave (1985).
Their results were plotted against the predictions of the SAFTE® model and are shown in Figure 4. All parameters within the model were set to the default values with the acrophase (peak of the 24-hr circadian rhythm) and start time as indicated in the legend. The SAFTE® Model predictions for the actual data are exceptionally good, accounting for 98% of the variance in the results of a series of accepted laboratory tests.
Often demanding military and civilian schedules provide less than the optimal eight hours of sleep a day for extended periods of time. A recent study of chronic sleep restriction conducted at the Walter Reed Army Institute of Research in cooperation with the Department of Transportation provided data on schedules of seven, five, and three hours of time in bed over seven days [1, 3]. The down regulation of the reservoir capacity incorporated in the current SAFTE® Model was able to predict both the performance degradation effects and rate of recovery from those schedules with an R2 of 0.94 [4].
A large study of 1400 railroad accidents sponsored by the Federal Railroad Administration has demonstrated that the SAFTE® Model is capable of predicting increases in human factors accident likelihood with an R2 of 0.93 [5]. Further analysis has indicated that the model also predicts increases in accident severity that accompany predicted fatigue (reduced effectiveness), with accident severity (total damage cost) significantly higher for accidents that occur when average effectiveness of the operators was below about 77%.
The SAFTE® model has received a broad scientific review and several services within the US Department of Defense consider it the most complete, accurate, and operationally practical model currently available to aid operator scheduling. At the Fatigue and Performance Modeling Workshop held in 2002, of the six fatigue models evaluated from around the world, the most recent version of SAFTE® had the lowest error of all models evaluated in terms of predicting the impact of chronic sleep restriction [6]. The model is routinely used by the Department of Transportation for evaluation of fatigue in rail and aviation operations.
Extrapolations to Performance of Operational Tasks
The sleep and performance model has been optimized to predict changes in cognitive capacity as measured by standard laboratory tests of cognitive performance. It is assumed that these tests measure changes in the fundamental capacity to perform a variety of tasks that rely, more or less, on the cognitive skills of discrimination, reaction time, mental processing, reasoning, and language comprehension and production. However, specific operational tasks vary in their reliance on these skills, and deficits in cognitive capacity may not produce identical reductions in the capacity to perform all tasks. It is reasonable to assume, however, that the changes in task performance would be correlated with changes in the underlying cognitive capacity. In other words, if one were to plot changes in task performance as a function of measured changes in cognitive capacity, there would be a monotonic relationship between the two variables. Therefore, if these two sets of data were available from a test population subjected to sleep deprivation, linear (or non-linear) regression techniques could be applied to derive a transform function; this transform translates predicted cognitive changes into changes in operational task performance. Based on this reasoning, the model can be extended to predict variations in any task or component of a task (given appropriate test data) using a generalized Task Effectiveness expression.
Click here to learn more
[1] Belenky, G.L., Wesensten, N.J., Thorne, D., Thomas, M., Sing, H., Redmond, D.P., Russo, M.B., and Balkin, T. (2003). Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: A sleep dose-response study. Journal of Sleep Research, 12, 1-12.
[2] Angus, R. and R. Heslegrave (1985). Effects of sleep loss on sustained cognitive performance during a command and control simulation. Behavior Research Methods, Instruments, and Computers, 17(1), 55-67.
[3] Balkin, T., Thorne, D., Sing, H., Thomas, M., Redmond, D., Wesensten, N., Williams, J., Hall, S., and Belenky, G. (2000). Effects of sleep schedules on commercial driver performance. (Report No. DOT-MC-00-133). Washington, DC: U.S. Department of Transportation, Federal Motor Carrier Safety Administration.
[4] Hursh, S.R., Redmond, D.P., Johnson, M.L., Thorne, D.R. Belenly, G., Balkin, T.J., Miller, J.C., Eddy, D.R., Storm, W.F. (2004). The DOD Sleep, Activity, Fatigue, and Task Effectiveness Model. Aviation Space and Environmental Medicine 75 (supplement 3, section II), A44-A53.
[5] Hursh, S.R., Raslear, T.G., Kaye, A.S., and Fanzone, J.F. (2006). Validation and calibration of a fatigue assessment tool for railroad work schedules, summary report (Report No. DOT/FRA/ORD-06/21). Washington, DC: U.S. Department of Transportation.
[6] Van Dongen, H. P. A. (2004). Comparison of Mathematical Model Predictions to Experimental Data of Fatigue and Performance. Aviation, Space and Environmental Medicine, 75, 3, 15-36.