Finding the inactivity threshold to accurately estimate computer use time with software

Details

• Personal Author:
  Chang CH ; Dennerlein, Jack T. ; Johnson PW
• Description:
  Aims: Using software which monitors computer keyboard and pointing device activities, this study determines the inactivity threshold that can accurately predict/identify the time someone working at the computer. With such threshold, computer work time can be accurately estimated. This study also determines if one general inactivity threshold can be used across individuals. While previous study tested only one single threshold, our study tested varied thresholds and assessed the variability of estimate errors across individuals to determine the optimal inactivity threshold. Methods: Twenty office workers, mostly administrative staff, who used computer more than half of their working hours were recruited for this field study. Eighteen subjects (90%) used desktop PC's for work and two used laptop PC's. For four consecutive hours, in each subject's own office, the keyboard and pointing device activity time were recorded by input device monitor software installed onto each subject's computer. Simultaneously, a video camera recorded the subject's interaction with the computer. For each subject, we calculated estimates of computer use time from the softwarerecorded data by cumulatively adding up input device activity time and pause time between successive activities that is shorter than a preset inactivity threshold. Twelve estimates were generated for each subject using twelve corresponding inactivity thresholds, 5, 10, 20, 30, 45, 60, 90, 120, 180, 240, 300 and 360 seconds. Video tapes were analyzed to determine the actual computer use time. In the video analysis, computer use time was defined as the time when the subject was using the keyboard, using the pointing device, hand poising above any input device, hand transferring between input devices or viewing the display monitor. The estimate errors were calculated as the difference between each of the twelve estimates and the actual computer use time calculated from the video tape (gold standard). We then calculated the average error and standard deviation of estimate error across subjects for each inactivity threshold. Data from ten subjects were analyzed at the time this abstract is submitted. Results: The average estimate error across subjects was the smallest when the inactivity threshold was set 45 or 60 seconds (-1% and +2%, respectively, Fig. 1). For individual subjects, the most accurate thresholds varied between 20 and 90 seconds. Although the standard deviation of estimate error was the smallest, 5%, when the inactivity threshold was set 90 seconds, those of 45 and 60 seconds (8% and 7%, respectively) were also close to 5%. The estimate error increased when the inactivity threshold deviated from 45-60 seconds; and the variability of error increased when the inactivity threshold deviated from 90 seconds. Conclusions: Preliminary results indicate that inactivity threshold between 45 and 60 seconds can accurately estimate computer usage time from software-recorded input device activities. With these two thresholds, the small variability of estimate error across subjects might also suggest that a general threshold can be applied on different individuals to estimate actual computer use time with sufficient accuracy. Such threshold will help implement computer activity monitoring software into field studies to objectively quantify the exposure duration of computer use. Further data analysis will increase the sample size for better generalizability and determine if the inactivity threshold changes by time within the same individual. [Description provided by NIOSH]
• Subjects:
  Computers Men Musculoskeletal System Occupational Health Safety Software Women
• Keywords:
  Computer-software Extremities Humans Monitoring-systems Monitors Muscle-function Musculoskeletal-system Statistical-analysis
• Publisher:
  International Commission on Occupational Health (ICOH)
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Abstract or Summary
• Place as Subject:
  Massachusetts ; OSHA Region 1 ; OSHA Region 10 ; Washington
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  13
• NIOSHTIC Number:
  nn:20046315
• Citation:
  Sixth International Scientific Conference on Prevention of Work-Related Musculoskeletal Disorders (PREMUS 2007), August 27-30, 2007 Boston, Massachusetts. Rome, Italy: International Commission on Occupational Health (ICOH), 2007 Aug; :13
• Federal Fiscal Year:
  2007
• Performing Organization:
  Harvard University, Boston, Massachusettts
• Peer Reviewed:
  False
• Start Date:
  19990930
• Source Full Name:
  Sixth International Scientific Conference on Prevention of Work-Related Musculoskeletal Disorders (PREMUS 2007), August 27-30, 2007 Boston, Massachusetts
• End Date:
  20090831
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:1eca77dccc43b6000ae11374bc3fe1ca26c7a4b2e3162d90cf755bc541b3ab640cde09bcca67968efdc0e514515e176ab8fff3701de083e401883aef1e09115e
• Download URL:
  https://stacks.cdc.gov/view/cdc/201412/cdc_201412_DS1.pdf
• File Type:
  [PDF - 48.31 KB ]