Using Bayesian Multilevel Modeling for Determining the Relationship between Shift work and Blood Pressure during a Retrospective longitudinal Study

Since there is inconsistency reports in relationship between shift work (SW) and blood pressure (BP), therefore we aimed to show any association between SW and BP by using of Bayesian Multilevel Modeling, which is a reliable method for this type of analysis. The profiles of 4145 workers in Polydactyl Iran Corporation were examined in historical cohort between 1996 until 2008. All relevant analysis was performed by Win Bugs software. Approximately 98 percent of study population was male. Of total 1886 (45.5%), 307(7.4%), 1952 (47.1%) of participation were day worker, two rotation shift worker and three rotation shift worker respectively. After controlling confounding factors, there was no significant relationship with Systolic BP (P=0.911) and Diastolic BP (P=0.278). In general, the results of our historical cohort study do not support a relationship between SW and BP. We suggest multi center and prospective cohort studies with controlling more confounding factors in this area.