Email updates

Keep up to date with the latest news and content from BMC Psychiatry and BioMed Central.

Open Access Research article

Suicide rates and income in São Paulo and Brazil: a temporal and spatial epidemiologic analysis from 1996 to 2008

Daniel H Bando1, Andre R Brunoni13, Isabela M Benseñor12 and Paulo A Lotufo124*

Author Affiliations

1 Doctoral Program of Sciences, Faculdade de Medicina, University of São Paulo, São Paulo, Brazil

2 Clinical and Epidemiological Research Center, Hospital Universitário, University of São Paulo, São Paulo, Brazil

3 Department of Neurosciences and Behavior, Instituto de Psicologia, University of São Paulo, São Paulo, Brazil

4 Clinical and Epidemiological Research Center, Hospital Universitário, Av Lineu Prestes 2565, 3° andar – Centro de Pesquisas Clínicas, Cidade Universitária, São Paulo, SP, Brazil

For all author emails, please log on.

BMC Psychiatry 2012, 12:127  doi:10.1186/1471-244X-12-127

Published: 28 August 2012



In a classical study, Durkheim noted a direct relation between suicide rates and wealth in the XIX century France. Since that time, several studies have verified this relationship. It is known that suicide rates are associated with income, although the direction of this association varies worldwide. Brazil presents a heterogeneous distribution of income and suicide across its territory; however, evaluation for an association between these variables has shown mixed results. We aimed to evaluate the relationship between suicide rates and income in Brazil, State of São Paulo (SP), and City of SP, considering geographical area and temporal trends.


Data were extracted from the National and State official statistics departments. Three socioeconomic areas were considered according to income, from the wealthiest (area 1) to the poorest (area 3). We also considered three regions: country-wide (27 Brazilian States and 558 Brazilian micro-regions), state-wide (645 counties of SP State), and city-wide (96 districts of SP city). Relative risks (RR) were calculated among areas 1, 2, and 3 for all regions, in a cross-sectional approach. Then, we used Joinpoint analysis to explore the temporal trends of suicide rates and SaTScan to investigate geographical clusters of high/low suicide rates across the territory.


Suicide rates in Brazil, the State of SP, and the city of SP were 6.2, 6.6, and 5.4 per 100,000, respectively. Taking suicide rates of the poorest area (3) as reference, the RR for the wealthiest area was 1.64, 0.88, and 1.65 for Brazil, State of SP, and city of SP, respectively (p for trend <0.05 for all analyses). Spatial cluster of high suicide rates were identified at Brazilian southern (RR = 2.37), state of SP western (RR = 1.32), and city of SP central (RR = 1.65) regions. A direct association between income and suicide were found for Brazil (OR = 2.59) and the city of SP (OR = 1.07), and an inverse association for the state of SP (OR = 0.49).


Temporospatial analyses revealed higher suicide rates in wealthier areas in Brazil and the city of SP and in poorer areas in the State of SP. We further discuss the role of socioeconomic characteristics for explaining these discrepancies and the importance of our findings in public health policies. Similar studies in other Brazilian States and developing countries are warranted.