Assessing agreement of self-reported and observed physical exposures of the upper extremity

Abstract

Assessment of workplace physical exposures by self-reported questionnaires has logistical advantages in population studies, but is subject to exposure misclassification. This study measured agreement between eight self-reported and observer-rated physical exposures to the hands and wrists, and evaluated predictors of intermethod agreement. Workers (n = 341) from three occupational categories (clerical/technical, construction, and service) completed self-administered questionnaires and worksite assessments. Analyses compared self-reported and observed ratings using a weighted kappa coefficient. Personal and psychosocial factors, presence of upper extremity symptoms, andjob type were evaluated as predictors of agreement. Weighted kappa values were substantial for lifting (0.67) and holding vibrating tools (0.61), moderate for forceful grip (0.58), and fair to poor for all other exposures. Upper extremity symptoms did not predict greater disagreement between self-reported and observed exposures. Occupational category was the only significant predictor of inter-method agreement. Self-reported exposures may provide a useful estimate of some work exposures for population studies.

Figure 1

Distribution of self-reported (solid lines) and observed (dashed lines) exposures for daily duration of time. Numbers on the horizontal axis represent time categories: none (1), less than five minutes (2), five to 30 minutes (3), more than 30 minutes but less than one hour (4), one to two hours (5), more than two hours but less than four hours (6), four or more hours (7) per day.

Figure 2

Distribution of self-reported exposures for three job types: clerical/technical (solid lines), construction (dashed lines), and service (dotted lines) worker groups. A) shows the results from the self-reported exposures; B) shows the results from the observed exposures. Numbers on the horizontal axis represent the same information as in Figure 1.

Figure 3

Prevalence of over-estimated and under-estimated values for self-reported and observed differences for each physical exposure by three job types. Percentage of over-estimated values are shown above the horizontal line and under-estimated values below.