Self-reported computer use duration biased the ezposure-response relationship of musculoskeletal symptoms
We completed a repeated-measure observational study to compare exposure-response relationships of musculoskeletal symptoms modeled on computer use duration assessed by self-report and a direct measure. For three one-week observational periods, 27 undergraduate students reported their daily computer use duration and upper extremity musculoskeletal symptoms to a handheld computer on average four times per day. Usage-monitor software installed onto each participants personal computer provided corresponding measures of daily computer use duration. For each handheld computer report, daily computer use duration was calculated from usage-monitor data within the time window starting from 6:00 AM in the same morning to the time when the report was completed. We used Mixed Effect Models to examine the relationship between daily computer use duration and the severity of musculoskeletal symptoms. Self-reported computer use and software-measured computer use were modeled separately and compared for their relationships with musculoskeletal symptoms. For software-measured computer use, every hour increase was related to a 34％ increase in the odds of reporting symptoms (odds ratio 1.34, 95％CI: 1.22-1.47, p=<0.01). For self-reported daily computer use duration, every hour increase was related to a 59％ increase in the odds of reporting one or more moderate or severe upper extremity symptoms (odds ratio 1.59, 95％CI: 1.24-2.32, p<0.01). When comparing daily computer use durations across symptom severity levels, both measurements exhibited a common pattern in which exposure duration increased from no symptom to moderate symptom, and then decreased at the level of severe symptom. Self-reported daily computer use was, however, generally 0.9-4.6 hours longer than the corresponding software-recorded value. In this pilot study of undergraduate students, self-reported daily computer use slightly biased the exposure-response relationship of musculoskeletal symptoms and inconsistently overestimated computer use. Researchers conducting epidemiologic studies need to be aware of the potential bias associated with self-reported exposure to computer use.
Che-hsu (Joe) Chang Cammie Chaumont Menéndez Michelle Robertson Benjamin C. Amick Ⅲ Peter W. Johnson Rosa J del Pino Jack T. Dennerlein
Harvard School ofPublic HealthBoston, MAUSA The National Institute forOccupational Safety and HealthMorgantown,WVUSA Liberty Mutual Research Institutefor Safety,Hopkinton, MA Institute for Work and HealthToronto, OntarioCanada University of WashingtonSeattle, WAUSA University of TexasHouston, TXUSA Harvard School of Public Health Boston, MA USA
国际会议
17th World Congress on Ergonomics(第十七届国际人类工效学大会)
北京
英文
1-5
2009-08-09（万方平台首次上网日期，不代表论文的发表时间）