Socioeconomic status, employment, migration and common mental disorders in Olinda, northeast Brazil

Abstract

A cross-sectional study was carried out in a low income area of Olinda, in the state of Pernambuco, Northeast Brazil, to investigate the specific links between social class and the prevalence of common mental disorders (CMD) assessed by the Self Reporting Questionnaire (SRQ-20). The analysis brought together a wide range of potential explanatory variables grouped under four headings: demographic characteristics, socioeconomic status (SES), employment and migration. Prior to adjustment, statistically significant associations were found between CMD and all SES variables (literacy, years of schooling, own occupation, income distribution, housing conditions, possession of household appliances), except housing tenure. After adjustment for all measures of SES, gender, age and marital status, CMD was more common in those with less education (X: for linear trend=10.37, df=l, p<O.OOOI) and low household per capita monthly income (X: for linear trend=5.21, df=l, p=0.02). Unemployment and informal work were independently associated with CMD among females (OR=2.66, 95% CI 1.1-6.3 and OR=3.02, 95% CI 1.3-7.2, respectively), but not among males (OR=1.77, 95% CI 0.8-3.9 and OR=1.08, 95% CI 0.5-2.4, respectively), though the interaction between employment status and gender was not significant. In the univariate analysis, migrants showed a higher prevalence of CMD than the native population. After controlling for gender and age, the association between migration and CMD was markedly reduced and became statistically non significant (OR=1.28, 95% CI 0.9-1.9). Validation of the SRQ-20 suggested that females, elderly, the less educated, housewives, migrants and those with the lowest income tended to over-report on the SRQ-20 compared to a psychiatric interview. These findings could be explained simply in terms of gender differences, suggesting that over-reporting was mostly related to being female. Education,representing early circumstances in life, seems to be the most powerful indicator of social class disadvantages in this area. Our findings suggest how the contribution of later socioeconomic conditions to the prevalence of CMD, such as occupation, housing conditions and household appliances, could be understood in part by their association with education. This is not to imply that later circumstances were reflecting only the influence of earlier life. Unemployed, informal workers and those belonging to low income households tended to be less educated. However, employment status and income showed an association with CMD independent to that of education.