Validity of the HSE stress tool: an investigation within four organizations by the Corporate Health and Performance Group

AIM: To examine the validity and accuracy of the HSE Management Standards Stress Tool. METHODS: A cross-sectional questionnaire was administered to employees of the occupational health and human resource departments of four of the Corporate Health and Performance Group organizations. The relationship between the six first-pass stress filter questions and self-rated health were examined using the General Health Questionnaire (GHQ12) and the Short Form. The relationship between the filter questions and objective work characteristics such as sickness absence, self-rated performance and job satisfaction was also examined. RESULTS: Two hundred and thirty-five employees participated, giving an overall response rate of 68%. The study sample met the proposed HSE of 85 and 65% pass rates and would not normally progress to further analysis. Almost 40% of the group were above threshold on the GHQ12; 19% rated their work as very or extremely stressful. While the filters were highly specific, they were insensitive and at best detected for less than 40% of those with psychological or work stress. All but one of the filters had less than 50% positive predictive value for work stress and the majority of those identified by the filters did not consider themselves stressed, either overall or by work. Five of the filter questions were significantly associated with self-rated performance, three with absence and two with other objective work indicators. CONCLUSIONS: Further developmental work on these filter questions is indicated. Use of work absence as a measure of the impact or 'cost' of stress may lead to a significant underestimation of the 'true cost' of psychosocial hazards in the workplace.