Modifying the Three-Process Model of Alertness for Accelerometer Data and Its Correlation with Sleep

• Activity Number: 124
• Type: Topic Contributed
• Date/Time: Monday, August 3, 2009 : 8:30 AM to 10:20 AM
• Sponsor: Section on Government Statistics
• Author(s): James Slaven and Ja K. Gu and Michael E. Andrew*+ and John M. Violanti and Luenda E. Charles and Bryan Vila and Cecil Burchfiel
• Companies: National Institute for Occupational Safety and Health and State University of New York at Buffalo and National Institute for Occupational Safety and Health and Washington State University and National Institute for Occupational Safety and Health
• Keywords: accelerometer ; alertness ; correlation analysis

Abstract

Alertness can be affected by poor sleep quality, from factors such as shift work. Assessing alertness to see whether it is associated with shift work requires a quantitative measurement, such as is given by Folkard and Akerstedt. This method has been adjusted to fit actigraphy data, rather than EEGs or sleep diaries. We used sleep data obtained from accelerometers, worn by police officers from Buffalo, NY, (BCOPS study) to show that actigraph data can be converted into alertness ratings using circadian rhythm and sleep/wake patterns. These ratings typically range from 3 (extreme sleepiness) to 14 (high alertness), with ratings below 7 considered as decreased alertness. The average alertness for police officers in our cohort is 6.07 (n=208). These derived alertness values correlate well with daily sleep duration (Pearson correlation coefficient = 0.9577, p < 0.0001).