Building Better Prediction Models for Exposure Assessment of Trapezius Muscle Activity During Computer Use

Details

• Personal Author:
  Bruno Garza JL ; Catalano PJ ; Dennerlein, Jack T. ; Eijckelhof BHW ; Huysmans MA ; Johnson PW ; Katz JN ; van der Beek AJ ; van Dieen JH
• Description:
  Purposes: Prediction models have been proposed as an alternative method of exposure assessment for epidemiologic studies that are currently difficult to perform within the field of ergonomics (Winkel and Mathiassen 1994, Svendsen et al 2005). Previous prediction models based only on task have been unsuccessful. We developed and evaluated a prediction model of trapezius muscle activity during computer use including both task and other predictors (individual characteristics, job characteristics, computer work behaviors, psychosocial factors, workstation setups, health outcomes, and leisure time exposures). Methods: We predicted median right and left trapezius muscle activities using 103 potential predictors of trapezius muscle activity during computer use. Trapezius muscle activities of 120 computer workers were calculated using data from Bruno Garza et al (2012). Potential predictors were categorized as individual characteristics, job characteristics, computer work behaviors, psychosocial factors, workstation setups, health outcomes, and leisure time exposures, and measured using a questionnaire (Ijmker et al 2006) and direct measurements of task, anthropometry, and workstation setup. Predictions were generated using a linear regression incorporating the relevant predictors as determined using a two-step backwards selection technique. All predictors had p<0.10 in the final model. The efficiency of predictions was evaluated using r-squared values and root mean squared (RMS) errors of the predicted compared to the measured values. We also calculated the percentage increase in participants needed in order to have the same power for epidemiological studies when using the model-based predictions, which is proportional to the increased standard deviation in the predicted dataset compared to the observed dataset, by dividing the square root of the R-squared value by the slope from the regression equation. Results: The r-squared values were 0.39 for the right trapezius and 0.52 for the left trapezius. The RMS errors were 2%MVC for both the right and left trapezius. The percent increase in number of participants needed was 255% for the right trapezius and 191% for the left trapezius. Task was significant in the right trapezius model only. Conclusions: We were able to explain a majority of the variation in left median trapezius muscle activity, and the RMS errors for both right and left trapezius were fairly small. More participants would be needed to use predictions than direct measurements, but the collection of the necessary information for predictions is likely less time and cost intensive than collection of direct measurements. Our models performed better than previous prediction models based only on task. [Description provided by NIOSH]
• Subjects:
  Computers Musculoskeletal System Occupational Health Risk Assessment Risk Factors Safety Workplace
• Keywords:
  Author Keywords: Exposure Assessment Computer Workers Models Muscle Function Musculoskeletal Psychosocial Factors Risk Factors Task-based Trapezius Upper Extremities Work Environment

  More +
• 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:
  223-224
• NIOSHTIC Number:
  nn:20056481
• Citation:
  Eighth International Scientific Conference on Prevention of Work-Related Musculoskeletal Disorders (PREMUS 2013), July 7-11, 2013 Busan, Korea. Rome, Italy: International Commission on Occupational Health (ICOH), 2013 Jul; :223-224
• Email:
  jlb612@mail.harvard.edu
• Federal Fiscal Year:
  2013
• NORA Priority Area:
  Healthcare and Social Assistance ; Services
• Performing Organization:
  Harvard University, School of Public Health, Boston, Massachusetts
• Peer Reviewed:
  False
• Start Date:
  20080901
• Source Full Name:
  Eighth International Scientific Conference on Prevention of Work-Related Musculoskeletal Disorders (PREMUS 2013), July 7-11, 2013 Busan, Korea
• End Date:
  20130831
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:3386efce2f0b889382bd8d673bb73a9f847cee74e88bf378ce751bfea1d5b5acdb583e1dd072e1ca2aedf41f838d995cda81f3f56f8de43d5af21373624a7883
• Download URL:
  https://stacks.cdc.gov/view/cdc/214084/cdc_214084_DS1.pdf
• File Type:
  [PDF - 721.11 KB ]