iWorkHealth: An instrument to identify workplace psychosocial risk factors for a multi-ethnic Asian working population

OBJECTIVE: The current study aimed to develop iWorkHealth, a valid and reliable self-administered instrument which identifies workplace psychosocial risk factors in Singapore. METHODS: The survey was conducted among 2718 employees who were primarily salaried workers and working in five companies from the healthcare, banking and finance, and legal sectors in Singapore. Factor extraction and item reduction were conducted using exploratory factor analysis (EFA) and Mokken scale analysis (MSA). Construct validity, internal consistency and convergent validity of the final scale were confirmed using confirmatory factor analysis (CFA), Cronbach's alpha and Pearson correlation coefficients, respectively. Multiple Indicators Multiple Causes model was used to detect Differential Item Functioning (DIF). RESULTS: EFA and MSA identified a five-factor solution (job demand, job control, employee and management engagement, supervisor support and colleague support) for the 27 items iWorkHealth instrument. CFA demonstrated that the five-factor model fitted the data with high internal consistency (Cronbach's alpha ranged from 0.79 to 0.92). The convergent validity was shown through significant association with existing scales—high job demand was significantly associated with high burnout and depression, while high job control, employee and management engagement, supervisor support and coworker support were significantly associated with low burnout and depression. Ten items were detected with significant DIF, but impact was minimal on the associations between socio-demographics factors and iWorkHealth subscales. CONCLUSIONS: The findings provided evidence that the iWorkHealth instrument which comprises 27 items in five domains of psychosocial risk at the workplace is a reliable and valid instrument that could be used to measure and compare the level of psychosocial risk factors across companies and industries in Singapore.