The study’s main findings evidenced the magnitude of the suicide problem in Brazilian society. Suicide has taken more lives in Brazil with each passing year and is frequent in all adult age brackets (starting at 20 years), but the rates have grown at particularly alarming rates in the elderly population. And although the suicide rates in women were four times lower than in men, younger Brazilian women have nevertheless increasingly died by suicide. There are also differences in the regional suicide mortality rates: while for many years the South of Brazil had the most cases and highest rates [28–31], it has shown a downward pattern in rates in the period studied and in the relative risk in the more recent cohorts. Meanwhile, Northeast Brazil has displayed alarming growth in suicide rates and the Southeast has witnessed an increasing risk of suicide in the most recent years analyzed.
This study aimed to estimate suicide trends over time in Brazilian society. The findings reinforce questions that had already been raised in other studies and shed light on possible suicide prevention strategies.
Age has always been aired as related to illness and death, and its effects have always been related to relevant health outcomes [20, 32], due to physiological changes, accumulated social experience, social role, changes in status, or a combination of the above. These changes thus reflect the biological and social process of individuals’ internal aging and represent changes in life-course development . Age is considered one of the main risk factors for suicide [33, 34], with the extreme age groups – the youngest adults [30, 33, 35–38] and the elderly [30, 34, 38–43] – as those most affected. Our study revealed a clear relationship between the increase in suicide rates over the course of life in practically all of the subsets tested. Only women 20–24 years of age in the North showed higher suicide rates than all the other age brackets.
We opted not to include individuals under 20 years of age, realizing that the impact on younger people (children and adolescents) has been small compared to the other age rates and that the inclusion of very extreme rates would hinder the analysis of the proposed modeling. However, women have died of suicide earlier than men, and this should be considered in any strategy to confront the problem. Suicide in young people is closely related to the life phase in which they are making choices that determine their fates and life projects  and that create intense economic impacts. Various factors have been listed as potential causes of suicide among young people and adolescents: unemployment, economic difficulties, family breakdown, and changes in society, such as decreased religiousness, new gender roles, increasing competition in school, increasing female presence in universities, and the shift from traditional rural society to an urban and industrialized structure . From 1980 to 2000, Brazil experienced an alarming 1,900% increase in suicide rates in the young population , corroborating our findings, although the South of Brazil showed a clear decrease in the rates in this same age bracket.
The increase in suicide rates among the elderly, as evidenced in this study, is consistent with findings in other countries and raises a serious concern, since the world population’s life expectancy is increasing, and along with it, the proportion of the elderly population in societies’ composition . In addition, the method used most frequently in this age group has been self-poisoning either with pesticides or medicines, and access to the latter is easier in this age bracket due to various clinical conditions that are more frequent in the elderly. Clearly, more elderly persons have shown greater risk due to two factors, age and generation. As for age, the reasons for developing suicidal ideation include the impossibility of coping with life as previously, loss of life companions, and disabilities resulting from illnesses or aging itself . The reality of many elderly people is rife with chronic illnesses, which can often cause emotional changes and decreased functional capacity. In addition, retiring from the world of work, loss of contact with coworkers, and the death of loved ones result in the lack of a social place and exacerbate the impacts of chronic illness .
This study showed that gender and age were the most important factors for explaining suicide rates in Brazil, as in the only other study with APC modeling in Brazil  and in other countries [9, 11–14].
The generation effect is even more serious but has received little attention. The cohort or generation effect is the change in groups’ experience in an initial event such as birth in the same year or ranges of years  and that undergo different exposures to socioeconomic, behavioral, and environmental risk factors in various life cycles . Brazilian society is undergoing an accelerated demographic transition in which personal relations have changed greatly: fertility has decreased, even as childbearing age has started earlier (with earlier sexual maturity) and has lasted longer due to medical technology that maintains biological viability of pregnancies in older women, as well as procedures that have extended men’s sexual life with medications; the nuclear family no longer has a perpetual configuration, marriage is no longer indissoluble, and relationships are quicker and less bureaucratic, without maintaining the patriarchal hierarchy of the last century; and despite social networks and the modern world’s dynamics, individuals feel increasingly isolated and removed from personal relations. This process of social change together with an older population (more physically, economically, and technologically limited) aggravates processes of isolation and depression, leading to suicidal ideation much more often than in other age groups. Our findings show that for practically all the regions and for Brazil as a whole, the more recent cohorts (1991 to 1995) evidenced higher risk when evaluating the overall population. The South was the only region of Brazil with higher suicide risk in the oldest cohort (1906 a 1910).
Other studies have largely pointed to greater importance of these effects on suicide rates, especially in relation to cohorts. In Switzerland, the cohort effects were similar for the male and female populations, although in the latter the effects were less pronounced, but were determinant for understanding the behavior of suicide rates in that country . A study in Spain found a period effect for the female population, while the cohort effect was more evident in the male population . In South Korea, cohort effects were determinant in the changes in rates from 1984 to 2013 . A study conducted in Hong Kong and Taiwan found that the age effects for both regions in both sexes were quite similar and suicide rates increased with increasing age. Regarding period effects, Hong Kong had one peak (1999–2003) and Taiwan had two peaks (1979–1983 and 2004–2008). As for cohort effects, in both Hong Kong and Taiwan, younger male cohorts showed high suicide risk; while younger female cohorts, however, showed relatively low risk . In the state of Rio de Janeiro, Brazil, from 1979 to 1998, the age-adjusted rates increased, more in men than in women, while a weak period effect was seen in the increasing rates in 1983–1984 and the cohort effect showed a decrease in the rates between the oldest cohorts and the youngest . However, in our study, the period effect was larger than the cohort effect on suicide rates for most of the situations analyzed. The cohort effect was only stronger for the overall Brazilian population, the overall and male populations in the Southeast, and the female populations in the North, Central-West, and South.
Period effect is the variation in the time period that affects all the groups simultaneously, representing the rate’s change in successive time periods [20, 33], summarizing a complex set of historical events and environmental factors, such as world wars, economic booms and recessions, famine, infectious disease pandemics, public health interventions, and technological discoveries . The literature includes reports on the association between suicide and economic crises that resulted in economic recession and unemployment in various regions of the world [14, 47, 48].
The specific Brazilian case has shown that the rates have grown over the years [42, 49–51], raising the possibility that economic status has been a strong factor for the problem, together with the trends in personal and social relations addressed above in this article. The Brazilian economy has alternated short growth cycles with economic slowdowns, generally abrupt, since the 1980s when the dictatorship came to an end and the country began its process of re-democratization. This growth pattern persisted over the early 2000s. In the year 2000, Brazil’s per capita Gross Domestic Product grew 4.3%, but in the following years (2001 to 2003) the economic slowdown had a heavy impact, and the GDP grew by only 1.7% per year. This economic scenario was followed by years of abundance and growth until mid-2014, except for a short period in 2009, when Brazil suffered the impact of the global crisis . Our findings showed that an increase in suicide risk in the period from 2001 to 2005 in the overall population of Brazil and in some regions suggested that this period of economic and social transition with changes in the political profile resulting in conditions of emotional instabilities that provided fertile ground for self-destructive behavior. This analysis is important, because the Brazilian economy entered another heavy and prolonged recession starting in 2015, starting with crises in the industrial and services sectors and still showing no signs of recovery . One can thus infer that the suicide rates since 2015 may have been influenced by this panorama and that feasible strategies need to be designed to minimize the harms from the economic impact on the Brazilian population’s mental health.
This finding on the period effect with the study in Spain showed that socioeconomic and structural changes were responsible for the increase in depression, alcoholism, and suicidal ideation in the 1980s . Likewise, in Switzerland, the two World Wars and economic problems were the factors impacting suicide rates in both sexes . In Russia, the Cold War, Mikail Gorbatchev’s plans, and the breakup of the Soviet Union were the backdrop for the increase in suicide rates among Russians . The concern with this effect’s impact is due mainly to doubts as to the repercussions of this scenario on future suicide rates, since Brazil is suffering the effects of the global economic crisis and globalization of consumption, leading to cheaper costs for importation than for domestic production, alongside the unfavorable political scenario stemming from the discovery of corruption schemes in large state-owned companies, drastic government changes, and dwindling investments in social programs. This combination of factors tends to aggravate the changes in society and predict an increase in psychological suffering, depression, and other mental disorders that can trigger suicides.
Except for the North and for women in the Central-West region, the model that best fits the data is the complete model (age-period-cohort). Table 1 shows that cohort effect had a stronger influence than period effect for the overall population of Brazil and the Central-West and Southeast, the male population of the Southeast, and the female population in the South. The period effect had a stronger influence in all the other cases.
A question that cannot be overlooked is that this increase in the absolute number of suicides and in the rates may result from improvement in the quality of mortality data. Confronting the idea of death mobilizes our own sense of finitude, an aspect that increases exponentially in the case of suicide . Suicide is still permeated by persistent taboos and prejudices, hindering the search for help in the presence of ideation, even though the problem is debated more widely today. Thus, part of the logical reasoning is to assume that underreporting can occur, although the official data on suicide are believed to be improving. However, admitting that a suicide has touched the individual’s family, the community, or society is still not a routine and demystified step. Admitting the suicide often means admitting that the psychological suffering existed and was not adequately treated to avoid resulting in the act, and that preventive policies have been insufficient. Still, admitting the fact is actually the first step towards changing the situation, as confirmed by the change experienced in the South of Brazil. Although the South of Brazil still has high suicide rates, our study showed that the younger cohorts have lower suicide risk and that the rates have decreased over time. This change has not appeared out of nowhere: it is the result of various studies [28–31] that reported higher suicidal behavior in this population, based on which, adequate suicide prevention policies were demanded that produced these new results over time.
The APC method also allowed detecting the differential magnitude of suicide’s impact between men and women. Not surprisingly, males predominated in the suicide profile, a fact that has been documented in other studies in Brazil [2, 4, 7, 8, 29, 30, 36, 49, 50, 54, 55] and elsewhere in the world [43, 56–61]. However, female suicide patterns call stand out, since women have tended to die of suicide at younger ages. Attempted suicide should also be considered, since it is more frequent in women [62–64]. The pattern may still be disguised by women’s choice of less effective and less violent methods . According to another theory, men generally tend to have more schooling than women, and women would thus display lower suicide rates. This difference in education has decreased over the years, but women have continued to be protected by having lower prevalence of alcohol abuse, greater religiousness, and flexibility in social skills and roles throughout life, while recognizing risk signs for depression earlier, participating more in social support networks [66–69], and seeking help for mental disorders earlier. Our findings evidenced that Brazilian women have displayed an upward pattern in suicide rates throughout adulthood, until around 50 years, when the rates decreased, except in the Northeast and Southeast regions, where they continued to increase.
Possible limitations are the type of data used (death certificate records) and uncertainty concerning this data’s quality. In addition, the study design (ecological) does not allow individualization of the findings, making it difficult to correlate the suicide event with the occurrence of depression and mental disorders in the Brazilian population and thus address possible causes of this problem.
On the issue of the method used here, APC modeling is not routine or simple to analyze. In other studies that use APC analysis of suicide data, there was no consensus on the best way to analyze the data [11–17]. The current study shares the limitations common to APC analysis. On the one hand, they are constrained by the lack of definitive real-world models, such that some ambiguity in the results cannot be ruled out. On the other, the effects of APC represent formal dimensions related to time and age, which are insignificant per se, but which provide useful tools for examining real-world variables. The analysis was also limited to the principal effects and leaves some room for more complex models, such as non-linear models or models including interaction effects.
Despite these possible limitations, age-period-cohort modeling is capable of describing suicide trends more accurately than other approaches. It obviously does not rule out the need for other epidemiological studies, but it focuses the attention where it has proven most urgent and where the interventions should begin.