The discovery of a significant suicide risk cluster represents an important point for further investigating the risk factors for suicide. In the case of the city of São Paulo, statistical analysis revealed that the relative risk of suicide is higher in the downtown, south-central and western-central areas of the city. Although the average rate for the entire city may be considered low (< 5.0 per 100,000 inhabitants) according to Diekstra and Gulbinat
], locally, the very downtown area of the city presented a rate of 6.3 per 100,000 inhabitants, a medium rate in this classification. This discovery warrants enhanced attention for public health action in these areas. It is important to mention that in the cluster detection procedure with a circular window in large areas with spatial variations in the population density, the resulting clusters can include districts with lower risk. Despite this limitation, in a recent review of software for space-time disease surveillance, SaTScan was highlighted as the most developed and robust software for cluster detection
Multiple logistic regressions pointed out that marital status plays an important role in suicide; even though estimates present wide confidence intervals, the results suggest that being single more than double the likelihood (OR
2.36) in high suicide risk clusters whereas being married almost halves such risk (OR
0.49). Durkheim was one of the first to observe that high suicide rates were associated with being single in France, and his theory could be used to explain this association in São Paulo. According to Durkheim’s theory, suicide varies inversely with the degree of integration of the individual to his social group. The spatial distribution of the proportion of singles (41.06%) and separated (18.61%) in the central region of the city of São Paulo is greater than the whole city (38.91% and 12.91%, respectively), which could lead one to interpret these suicides as similar to ‘selfish suicides’ in Durkheim’s work. Recent study based on suicidal individuals from São Paulo, from 1996–2008, confirmed that being single is significant as risk factor for suicide
In a review of 84 papers from 1981 to 1995, Stack
] concluded that more than three quarters of the studies supported the protective factor of being married. This was in agreement with his previous analysis of 15 nations (mostly European) in which he found that marriage lowered suicides for both males and females
]. A study of 12 developed countries with data from the end of the twentieth century, including the United States, Australia and European countries, identified marriage as suicide protective factor
]. The same effect was observed in Italy with data from 2000 to 2002
Data from Unites States between 1986 and 2002 showed that larger families were associated with lower suicide rates whereas divorced or separated, widowed, or never married individuals conditions were associated with higher suicide rates though only among men
]. In Northern Ireland, from 1996 to 2005, marriage was found to protect both genders against suicide
]. Another study verified protection only for married men in Austria using data from 1970 to 2001
]. A study with recent Taiwan data observed that never married males had the highest suicide rates
]. Divorce was also related to suicides in Japan
] and Romania
The migrant condition was found to increase the likelihood in high suicide risk clusters in almost 50% (OR
1.50) of São Paulo. The WHO’s “Self-directed violence” report
] alerts that suicide rates in a given migrant group have been found to be similar to that of the migrant’s country-of-birth. Indeed, Voracek et al.
] retrieved a meta-analysis of 33 studies with data from seven host countries, and a strong and direct association was found between immigrant rates and country-of-birth rates. The city of São Paulo still attracts a great number of migrants from all Brazilian regions
] whereas migration from other countries is also rising. In the downtown area, in the high suicide risk region of the city, migrants correspond to 79.8% of residents in contrast with only 13.2% of migrants in the city as a whole. Unfortunately, no information regarding the migrants’ origins is available to provide a control for this association. Whether this is due to imported patterns or to locally developed behaviours, the migrant condition may be a flag for suicide risk, which should at least invite health surveillance.
The Catholic condition was found to increase the likelihood in high suicide risk clusters in 37% (OR
1.37). Though at first this seems to conflict with Durkheim’s original finding that Protestants were more prone to suicide, one must understand that he attributed this finding to Protestants’ being less socially integrated. In São Paulo, the Catholic group is known to be less socially cohesive than Evangelical groups
] and thus may be susceptible to a higher suicide risk. In recent decades, the national census has shown a decrease in the number of Catholics in São Paulo (78.9 to 67.1%) and an increase of Evangelicals (8.54 to 16.4%)
]. Evangelicals have greater popular appeal. The number of Evangelical churches is three times the Catholic, in poor areas of the outskirts. Among the Evangelicals, there are several social networks that reduce their vulnerabilities
], thereby decreasing the suicide rates. Nevertheless, Prandi et al.
] showed a relationship between Evangelicals and poverty and found that the geographic distribution of these two variables was concentrated in the periphery of the city. In the state of Rio de Janeiro, Brazil, a recent ecological study found that being Evangelical is a protective factor for suicide
Neither income itself nor its putative interaction with religion achieved statistical significance in multiple logistic regression (Table
); therefore, only the crude protective effect of being Evangelical against suicide risk was detected. Regarding the results in Tables
, it may be possible to reconcile the observations found in the present study and in the one by Prandi et al.
]. The lower the income or the higher the proportion of Evangelicals, the less the risk (Table
In the present study, the chances of a high suicide risk cluster increased by 5.6% with each increase of one minimal wage level in income. In northern India, from 1996–2005, income was inversely related to suicide
]. Another study using World Health Organization (WHO) data focused on countries with a medium Human Development Index (HDI) and found that education and telephone density were directly related to suicide while a high Gini index was inversely related to suicide
]. Other recent ecological studies performed in the United States
], Northern England
] and Tuscany (Italy)
] identified an inverse relationship between income and suicide. Another study carried out in Italy suggested that regions with higher economic status have the highest suicide rates
]. Nevertheless, the relationship between suicide and wealth is far more complex because it relates to multiple socio-economic factors resulting from numerous historical and cultural elements inherent in each population. Accordingly, there are multiple divergent findings in the literature regarding the direction of this association (e.g., direct, inverse or no relation) between wealth and suicide. The manual of suicide prevention by PAHO/WHO et al.
] considers the extremes of economic strata as suicide risk factors. Many factors can contribute to this wide variation of findings including the size of the aggregated population, metrics of socio-economic characteristics and the inclusion of different potentially confounding variables, study designs, secular trends, and cultural aspects
The present study has some limitations. The first is inherent to the ecological study design. The association observed between variables at the group level does not necessarily represent the association that exists at the individual level. This bias is known as ecological fallacy
. Other limitation is related to data collection due to misclassification of suicide. Different procedures and cultural and social practices and values probably have various effects on death records and lead to misclassification of suicide
]. Among those is the fact that health insurance does not cover health expenses when injury or death is resulted of suicide attempt, which can lead to underestimation.
The results herein presented are in agreement with Durkheim’s original suggestion that poverty had a protective effect against suicide
]. This could suggest that economic development leads to individualism that, in turn, leads to higher suicide rates
]. Thus, to better understand the role of wealth, one should seek a better understanding of poverty as suggested by Paugam
], whose work was based on Castel’s
] work distinguishing between disqualifying poverty and integrated poverty. The latter, which best suits São Paulo, is found in societies where low life standards are compensated by a solidarity response within family, neighbourhood, and region.