Evaluation of Alternate Category Structures for the Strain Index: An Empirical Analysis

OBJECTIVE The purpose of this study was to develop alternative Strain Index risk classification categories. BACKGROUND Strain Index scores are usually categorized into four Strain Index "risk categories." The "original" risk categories were developed in the meat-packing industry and may not be fully applicable to other industries. METHOD Daily Strain Index scores were estimated among 276 manufacturing workers participating in a cohort study of occupational risk factors for hand-arm musculoskeletal symptoms. Each score was categorized using the original method and a new method based on quartiles of Strain Index score values among symptomatic participants. Models examining associations between original Strain Index risk categories and incident hand-arm symptoms were compared to models examining associations between the alternative Strain Index risk categories and incident hand-arm symptoms. RESULTS Compared to the respective referent categories, a twofold or greater increase in the risk of incident hand-arm symptoms was observed for the highest original Strain Index risk category (HR = 2.06, 95% CI = [1.08-3.92]) and for the second highest alternate Strain Index risk exposure category (HR = 2.21, 95% CI = [1.26-3.85]). Although significant associations between Strain Index risk category and incident hand-arm symptoms were observed for both Strain Index categorization methods, model fit statistics favored the alternate approach. CONCLUSION Results from this study suggests that the Strain Index risk category structure may need to be tailored to specific populations. APPLICATION If verified, results from this study provide a better way to identify hazardous manufacturing jobs and target them for exposure reduction.