Estresse no trabalho e pressão arterial: reflexões metodológicas sobre linearidade e operacionalização da exposição

Abstract

The demand-control model proposed by Karasek characterizes situations of stress in the workplace, that is defined by job strain resulting from a combination of the effect of psychological demand at work (stressors) and moderators of stress, especially decision authority (decision latitude). There are several ways to operationalize the exposure (job strain) described in the literature, however, there is no consensus about which is the best way to do it. Another issue often neglected, by most authors, is the shape of the exposure-response function in order to explore non-linearity of the association. Hypertension is one of the main cardiovascular disease assessed in studies on the effects of job strain. However, conflicting results reported on the association between job strain and hypertension. The aim of this paper is the study the association between job strain and increased blood pressure/hypertension using different forms of operationalization of stress at work and evaluate the curve of exposure-response regarding the linearity. The work is a sectional study developed in the Pró-Saúde population which consists of public servants of a university in Rio de Janeiro. Regression models using different distributions were used to estimate the association between job strain and blood pressure/hypertension. Nonlinear functions were used to investigate the assumption of linearity of the association between exposure and the outcomes. The results suggest that the relationship between dimension decision latitude and blood pressure is not linear. Significant associations were found between job strain and systolic and diastolic blood pressure for some forms of operationalization of exposure, when analyzed separately. However, when analyzed jointly by means of bivariate response regression models no association was found. Also it was not found association between job strain and hypertension for any form of operationalization of the exposure. The best statistical models were those in which the exposure was operationalizated by means quadrants determined by the median of each of the dimensions and the interaction term between psychological demand and decision latitude. In this study, these were the most adequate way of operationalization. The results also show that the association between the dimension decision latitude and blood pressure is not linear. In conclusion, there is no guarantee of linearity of the association between job strain and blood pressure and that this association can be influenced by the way of operationalization of exposure and outcome or modeling strategy.