IJmker Stefan, Mikkers Janneke, Blatter Birgitte M, van der Beek Allard J, van Mechelen Willem, Bongers Paulien M
Body@Work TNO VUmc, Research Centre Physical Activity, Work and Health, The Netherlands.
Appl Ergon. 2008 Nov;39(6):685-96. doi: 10.1016/j.apergo.2007.12.003. Epub 2008 Feb 4.
"Ergonomic" questionnaires are widely used in epidemiological field studies to study the association between workstation characteristics, work posture and musculoskeletal disorders among office workers. Findings have been inconsistent regarding the putative adverse effect of work postures. Underestimation of the true association might be present in studies due to misclassification of subjects to risk (i.e. exposed to non-neutral working postures) and no-risk categories (i.e. not exposed to non-neutral working postures) based on questionnaire responses. The objective of this study was to estimate the amount of misclassification resulting from the use of questionnaires.
Test-retest reliability and concurrent validity of a newly developed questionnaire was assessed. This questionnaire collects data on workstation characteristics and on individual characteristics during computer work (i.e. work postures, movements and habits). Pictures were added where possible to provide visual guidance. The study population consisted of 84 office workers of a research department. They filled out the questionnaire on the Internet twice, with an in-between period of 2 weeks. For a subgroup of workers (n=38), additional on-site observations and multiple manual goniometer measurements were performed.
Percentage agreement ranged between 71% and 100% for the test-retest analysis, between 31% and 100% for the comparison between questionnaire and on-site observation, and between 26% and 71% for the comparison between questionnaire and manual goniometer measurements. For 9 out of 12 tested items, the percentage agreement between questionnaire and manual goniometer measurements was below 50%.
The questionnaire collects reliable data on workstation characteristics and some individual characteristics during computer work (i.e. work movements and habits), but does not seem to be useful to collect data on work postures during computer work in epidemiological field studies among office workers.
“人体工程学”问卷在流行病学现场研究中被广泛用于研究办公室职员的工作场所特征、工作姿势与肌肉骨骼疾病之间的关联。关于工作姿势的假定不良影响，研究结果并不一致。由于根据问卷回答将受试者错误分类到风险组（即暴露于非中立工作姿势）和无风险组（即未暴露于非中立工作姿势），研究中可能存在对真实关联的低估。本研究的目的是估计使用问卷导致的错误分类量。
评估了一份新开发问卷的重测信度和同时效度。该问卷收集有关工作场所特征以及计算机工作期间个人特征（即工作姿势、动作和习惯）的数据。尽可能添加了图片以提供视觉指导。研究人群包括一个研究部门的84名办公室职员。他们在互联网上两次填写问卷，中间间隔2周。对一组工人（n = 38）进行了额外的现场观察和多次手动测角仪测量。
重测分析的百分比一致性在71%至100%之间，问卷与现场观察比较的百分比一致性在31%至100%之间，问卷与手动测角仪测量比较的百分比一致性在26%至71%之间。在12个测试项目中的9个项目中，问卷与手动测角仪测量之间的百分比一致性低于50%。
该问卷收集了关于工作场所特征以及计算机工作期间一些个人特征（即工作动作和习惯）的可靠数据，但在办公室职员的流行病学现场研究中，似乎无法用于收集计算机工作期间的工作姿势数据。
