Abstract Details

Title: Measuring Organizational Policies and Practices for Health and Safety and Disability Prevention
Authors: Sheilah Hogg-Johnson; Colette Severin; Selahadin Ibrahim; Michael Swift; Lynda Robson; Ivan Steenstra; Peter Smith; Benjamin Amick
Session: The Organizational level: disability and occupational health and safety management - St. Patrick, 17 Oct 2016, 3:45 PM - 5:00 PM

Background & Objectives

Workplace parties and occupational health & safety (OHS) professionals are increasingly interested in assessing OHS performance using leading indicators rather than lagging indicators such as injury rates. The study objective was to evaluate the psychometric performance of the Organizational Policies and Practices (OPP) instrument, including structural validity, reliability, predictive validity.

Methods

Persons most knowledgeable about OHS in Ontario workplaces from ten sectors were recruited to complete a survey including the 27 item OPP for their workplace. A subset of respondents repeated the survey two weeks later. The 27 items yield five subscales, Safety Practices (SP), Active Safety Leadership (ASL), Ergonomics Practices (EP), Disability Prevention (DP), and People Oriented Culture (POC). Subscales were evaluated for internal consistency and test-retest reliability. Confirmatory factor analysis (CFA) was used to assess structural validity. Survey data were linked to workers’ compensation injury claims for the two years following survey administration to assess predictive validity using overall claim rate, claim rate for musculoskeletal disorders (MSD) and disability duration for MSDs as outcomes with negative binomial and Cox regression models controlling for sector, size and region.

Results

Completed surveys were obtained from 1639 workplaces with repeated surveys for 191. Cronbach’s a for the five subscales showed good internal consistency (0.84 – 0.92). Test-retest reliability ICCs were SP 0.77, ASL 0.69, EP 0.67, DP 0.82, POC 0.80 and considered adequate when above 0.70. The five subscale structure was confirmed with excellent CFA goodness of fit (RMSEA 0.041, CRI 0.95, TLI 0.94). Higher values of SP, EP and POC significantly predict lower overall claim rates. Higher values of EP marginally and DP significantly predict lower MSD claim rates. Higher values of EP and DP are significantly associated with faster times to return to work.

Conclusions

Psychometric properties of the OPP subscales are promising. Further investigation of validity is warranted.