Predictive validity of the Strain Index in turkey processing

The Strain Index is a job analysis method for determining if workers are ex posed to increased risk of developing distal upper extremity disorders. Its predictive and external validity was initially demonstrated in a pork proc essing plant. The purpose of this study was to evaluate the predictive vali dity of the Strain Index in one turkey processing plant. While blinded to h ealth outcomes, investigators analyzed the right and left sides of workers in 28 jobs using the Strain Index and classified them as "hazardous" or ''s afe" based on the Strain Index score. Subsequently, OSHA 200 logs were used to ascertain the occurrence of distal upper extremity disorders retrospect ively. If at least one such disorder had occurred on the right or left side during the previous 3 years, that side was classified as "positive." If no such disorder was reported during the previous 3 years, that side was clas sified as "negative. When comparing sides, symmetry between morbidity and h azard classification was required. When comparing jobs, such symmetry was n ot required. Evidence of association between the hazard classifications and the morbidity classifications for the 56 sides and the 28 jobs was evaluat ed using 2 x 2 contingency tables. Ebr the sides, the association between h azard classification and morbidity, classification was statistically signif icant, with an odds ratio of 22.0. The sensitivity, specificity, positive p redictive value, and negative predictive value were 0.86, 0.79, 0.92, and 0 .65, respectively. Similar results were noted for the jobs-the odds ratio w as 50.0 and the sensitivity, specificity, positive predictive value, and ne gative predictive value were 0.91, 0.83, 0.95, and 0.71. These results prov ide additional evidence of the external validity and predictive validity of the Strain Index.