Validitet och reliabilitet av ett arbetsbelastningsindex (AI) i en FHV enkät

Sammanfattning: Abstract   The purpose of this work was to scientifically evaluate the sustainability of parts of  a company-specific survey within the Occupational Health Services (FHV). The service is called HALU which is an abbreviation of the “Work Health and Lifestyle” survey and provided by Feelgood AB. HALU is a questionnaire that is used frequently towards the customer and answered on an individual level with subsequent health interviews and a statement regarding the organization's performance in order to identify the health, lifestyle and working situation of the organization. In this work, questions regarding workloads were evaluated. There were six different workload questions, and the question in this paper were whether these issues had an internal consistency, validity and reliability, and if these issues’ correlation to questions about health during the past year. These analyzes were designed as cross-sectional studies The work also included a longitudinal study in which survey responses from the same individual on two occasions analyzed to see if the responses of workload and health has correlation at two different response times. The material used is a selection from Feelgood database from which includes total includes 130,422 HALU surveys. This sample includes 17,513 surveys from 201007. The study was performed in three steps: Step 1 was a reliability study in which the internal consistency analysis regarding workload questions in the questionnaire . A workload index was created (AI) Stage 2 was a validity study where the correlation coefficient analyzed for correlation between the AI which was calculated in step 1 and the in -depth study of health issues. Step 3 was also a validity study where the correlation coefficient was analyzed in terms of relationships between different survey occasions in the same individual, in order to see if the change in workload results in change in self-perceived health . The result showed that there was strong internal consistency regarding workload issues and that there was correlation between workload and health in a cross section . There were also correlation between survey responses occasions but this was low.