Suicide in the elderly : approach to social determinants of health in the Dahlgren and Whitehead model

e20200332 http://dx.doi.org/10.1590/0034-7167-2020-0332 9 of ABSTRACT Objective: Identify in literature the social determinants of health related to suicide in the elderly, according to the model proposed by Dahlgren and Whitehead. Method: Integrative review of articles indexed in the databases BDENF, CINAHL, LILACS, and MEDLINE, with the following main descriptors: aged, suicide, social determinants of health, and risk factors. Primary studies were included which addressed social determinants of health and suicide in the elderly. Results: From the 19 articles analyzed, three categories emerged: proximal social determinants of health (male gender, mental disorders, physical illnesses, white race, 70-74 years old); intermediate social determinants of health (substance abuse, use of alcohol or psychotropic drugs, marital status, marital, social, and family problems, violence, previous suicide attempt, history of admission to psychiatric service); and distal social determinants of health (schooling, economic issues, sanitation, stressful events). Conclusion: Proximal determinants have more effects on suicide. Intermediate determinants are composed mainly of changeable factors. Distal determinants showed lesser associations. Descriptors: Suicide; Elderly; Social Determinants of Health; Risk Factors; Mental Health.


INTRODUCTION
Aging means going through a stage of human development that encompasses the interaction of biological, psychological, and social aspects, particular to the life of elderly people.The loss of autonomy as well as the appearance of pathologies and physical restrictions generate feelings of dissatisfaction and uselessness, leading to a psychological discomfort that implies a reduction in the quality of life.These changes in the course of aging are risk factors for suicide (1) .This is a complex phenomenon, influenced by multiple variables, which causes intentionally self-inflicted death (2) .It is estimated that each year 800 thousand people commit suicide -and age is a strong aspect of its occurrence.Higher mortality rates have been observed among individuals aged 70 or older compared to younger people (3) .
There is no singular motivation for this act.I may be a result of an interaction of factors.Unlike what occurs in other age groups, in the elderly it is observed that suicidal behavior has manifestations that are difficult to verify.However, one thing is certain: methods are more lethal and passive suicide is frequent, for instance when refusing food (4) .
This scenario points to a need of understanding the determinants for the occurrence of suicide in the elderly, especially considering that those are commonly related to a wide network of social and cultural aspects of the individuals themselves and their families (1,5) .
The social determinants of health (SDH) are the circumstances in which individuals are born, develop, live, work, and age.They are outlined by social, economic, cultural, psychological, and behavioral arrangements, which have an influence on the recurrence of health disorders and risk conditions in the population.This set of elements must be organized and explored, in order to define the singularity of subjects and the particularity of phenomenons (4,6) .
Identifying the SDHs that are related to suicidal acts among the elderly has implications for the unique context of each individual, based on their visibility, especially by health professionals.With this understanding, the determinants will be able to be tracked and monitored in the scope of assistance.For the social panorama, the understanding of the SDHs involved in each phenomenon and by age group can reduce the disparity between groups and greatly strengthen equity in health.

OBJECTIVE
To identify in literature the social determinants of health related to suicide in the elderly according to the model proposed by Dahlgren  and Whitehead.

METHODS
This is an integrative literature review.Its construction was carried out following the steps proposed by Ganong (7) : formulation of the problem and elaboration of the guiding question; selection of articles; categorization of studies; analysis of the data obtained; discussion and interpretation of results; presentation of the integrative review and synthesis of knowledge.
Initially, it was verified the available knowledge about the relationship between the social determinants of health and the occurrence of suicide in the elderly.A research question was therefore elaborated according to the PICo strategy, in which "P" corresponds to Participants, "I" to phenomenon of Interest and "Co" to Context of the study (8) .The following question came out: what are the social determinants of health associated with the occurrence of suicide in the elderly population?
The controlled (CD) and uncontrolled (UCD) descriptors were then selected.To do so, it were consulted Health Sciences Descriptors (DeCS in Portuguese), Medical Subject Headings (MeSH) and List of Headings of CINAHL Information Systems, which are presented in Chart 1.
The bases selected for this search were: Nursing Database (BDENF), Index to Nursing and Allied Health Literature (CINAHL), Latin American and Caribbean Literature in Health Sciences (LILACS) and Medical Literature Analysis and Retrieval System Online (MEDLINE), via PubMed.The search took place between June and August 2019, by two researchers, simultaneously, in the aforementioned databases, accessed through the Periodical Portal of the Coordination for the Improvement of Higher Education Personnel (CAPES in Portuguese), in an area with Internet Protocol (IP) recognized.The inclusion criteria were: articles (primary studies) that addressed the social determinants of health and the occurrence of suicide in the elderly population.Studies that addressed ideation and/or suicide attempt or that did not answer the research question were excluded.It is noteworthy that there was no time or language cut to select publications, for the objective was actually achieve a comprehensive approach.
The selection of works started by reading titles and abstracts, based on the inclusion criteria.Subsequently, a comprehensive analysis of the selected articles was performed and, to extract the data of interest, a validated form was used, with the following items: definition of subjects, methodology, sample size, measurement of variables, analysis method and supporting concepts employed.The level of evidence was classified as established by the Collaborating Center of the Joanna Briggs Institute (JBI) (9) .
The studies were grouped according to the similarity between objectives, results, and conclusion; they were organized into thematic categories following the model proposed by Dahlgren and Whitehead (10) .The model shows that the SDHs are allocated in different layers, from a closer one, included in the individual determinants, to a more distant one, in which the macrodeterminants (6) are found.
Thus, the theoretical categories listed were: proximal determinants, those related to the individual (age, sex, heredity and ethnic/racial conditions); intermediate determinants, which are allusive to lifestyle, social and community media; and distal determinants, which address living, working, and socioeconomic, cultural and environmental conditions in general (10) .

RESULTS
After reading titles and abstracts, 87 articles were obtained for full reading.Of these, 21 were considered potentially eligible.After analysis, 19 articles were included in the final sample of this integrative review (Figure 1), which followed Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA)'s recommendations.
Chart 2 shows the characterization of all 19 articles used in this review.Regarding the study sites, most were carried out in the United States (USA).The years of publication vary between 1994 and 2019.As for the design, most were cross-sectional (12 studies), with a level of evidence III 3. The SDHs found in each study are distributed in the following chart: Chart 2 -Characterization of selected studies according to authors, country, year of publication, study design, number of participants, objectives, social determinants of health, and level of evidence, Teresina, Piauí, Brazil, 2019

Autors
Country / Year of publication

Social determinants of health (SDH) NE
Cavalcante FG, Minayo MCS (11) Brazil / 2012 Descriptive study with a qualitative approach N = 51 Performing at least 50 psychosocial autopsies with relatives of elderly who had died by suicide.That in ten municipalities with high rates of these events, spread through all five Brazilian regions.

Proximal SDHs: Male sex, physical illnesses, and mental disorders
Intermediate SDHs: Marital status, social problems, previous suicide attempt Distal SDHs: Schooling, economic issues III 3 Minayo MCS, Cavalcante FG (12) Brazil / 2013 Descriptive study with a qualitative approach N = 11 Empirically understanding the cases and multiple causes of 11 women who died by suicide and the fragments of personal stories that led to self-inflicted death, told by their families.

Social determinants of health (SDH) NE
Cheung G, Douwes G, Sundram F (22) New Zealand / 2017 Cross-sectional and retrospective study N = 214 Comparing the sociodemographic and clinical characteristics of elderly with and without terminal cancer who died by suicide and analyze the suicide reasons of those with terminal cancer to determine whether they represent rational suicide.

Proximal SDHs: Male sex and mental disorder
Intermediate SDHs: Previous suicide attempt III 2 Ho RCM, Ho ECL, Tai BC, Ng WY, Chia BH (23) Singapore / 2014 Cross-sectional study N = 409 Identifying clinical predictor factors of suicide in the elderly, especially those without a history of suicidal behavior.

Proximal SDHs: Mental disorder
Intermediate SDHs: Social problems, alcohol consumption, use of psychotropic drugs III 2 Carney SS, Rich CL, Burke PA, Fowler RC (24) United States / 1994 Cross-sectional study N = 204 Verifying whether selected clinical factors that characterize suicides after 60 years old differ from factors in younger suicidal people and determine characteristics of male and female suicide in the elderly.
Intermediate SDHs: Marital status and previous suicide attempt Préville M, Hébert R, Boyer R, Bravo G, Seguin M (25) Canada
From a gender perspective, some studies that addressed suicide in the elderly population have demonstrated its association with males (11,13,(16)(17)(18)(19)22,29) . The lierature indicates that, although women have higher rates of suicide attempts, men actually do so, being responsible for the highest number of suicides.This is mainly attributed to the methods used, which have a higher lethality level (31) .However, other explanations for this finding call into question the socially superior masculinity model as a vulnerability factor for suicide among elderly men, mostly due to the difficulty of performing cultural roles that are imposed (32) .
In third age, men tend to value stoicism, control of emotions, sexism, and competitiveness.The role of family head and provider, of Suicide in the elderly: approach to social determinants of health in the Dahlgren and Whitehead model Carvalho ML, Costa APC, Monteiro CFS, Figueiredo MLF, Avelino FVSD, Rocha SS.
formerly belonging to the elderly, begins to be played by others, causing him to feel useless, incapable, and powerless.Leaving work because of retirement, chronic illnesses, financial problems, relationship problems, and changes in sexual performance are also potential aspects for suicidal behavior in the elderly (32) .Another factor raised through the analysis of the articles was the presence of mental disorder (11,13,(15)(16)(17)(18)(19)(22)(23)(25)(26)(28)(29) , the most mentioned disorder being depression and the others listed as bipolar disorder, anxiety and adjustment disorder. The presence of pschological disorder appears as an important vulnerabilizing aspect for suicide (33) .It is assumed that 90% of people who commit suicide have disorders of this sort (34) .However, depression occupies a prominent position, being considered the most common disorder in individuals who die by suicide (35) .
It is worth mention that the elderly are more resistant to show depressive symptoms or suicidal ideation.Thus, few receive diagnosis, and even fewer have access to treatment (36) .It is at this stage of life that depression shows its worse prognosis and a higher incidence of suicides (37) .
In addition to depression, other disorders are reported in literature as risk factors for self-inflicted death.A study conducted with 410 patients from a Psychosocial Care Center in Minas Gerais observed a statistically significant association between suicidal behavior and psychiatric comorbidity (psychotic or delusional disorders and maniac episode or bipolar affective disorder) among women.Men, in turn, presented a significant relationship between depression and psychiatric comorbidity (depression, schizophrenia, or disorders due to illicit drug use) and suicide (34) .
An integrative review with a time frame from 2002 to 2017, which evaluated ideation and suicide attempt in the elderly hospitalized in long-term care institutions, also mentions psychiatric problems as an important risk factor for the elderly population in general, highlighting depression, illicit substance abuse, personality disorders, behavioral disorder, previous history of suicide attempts, self-destructive indirect behaviors, direct harmful behaviors, cognitive impairment, and previous psychiatric diagnosis (38) .
Another risk factor for self-inflicted violence observed among the studies was the presence of physical disease (11)(12)19,27) . A foregn study shows that the number of diseases increases the cumulative risk of suicide.Elderly people are six times more likely to acquire some physical disease when compared to young people, and are more vulnerable to multiple comorbidities, use of different medications and functional decline, aspects that make them even more prone to the risk of suicide (39) .
In the sociocultural panorama, the value of an individual corresponds to his productivity.The decrease in the functional capacity of elderly people can make them be seen as useless in some situations and awaken feelings of sadness and anguish.In addition, the elderly tend to self-judge as a hindrance, a situation that can be aggravated by the loss of autonomy and dependence on basic care (40) .
With regard to race, the white race was pointed out as a risk factor (15,18) .Epidemiological studies that address this theme corroborate the finding and demonstrate that there is a higher occurrence of suicide in white than non-white people (41) .Research carried out at a national level found that, of the 164 cases of suicide attempt included, 46.3% were of the white race (42) .
The age group of 70 to 74 years old was also identified as more prone to suicide (20) .In a study on self-inflicted death in the elderly, carried out in the state of Piauí, the most prevalent age group was between 60 and 70 (54% of the sample), followed by 71 to 80 (34.5% of the sample).Statistical data also reveal that there is a tendency for the suicide rate to increase in the population aged between 70 and 80 years old (1) .
The analysis of the proximal SDHs considered in the present review reinforces the need to pay attention to individual characteristics in each elderly person, especially for the care established throughtout the territory, since these factors are intrinsic to the subject.Primary Health Care, in this context, can be taken as a differential device, which easily identifies such aspects and projects comprehensive assistance, including, beyond observing these SDHs, interventions toward risky behaviors.

II -Intermediate social determinants of health
In the second category, which addresses the middle layer of the SDHs, 14 articles referred to the following factors: substance abuse, alcohol consumption, use of psychotropic drugs, marital status, marital problems, social and family problems, violence, previous suicide attempt, and history of admission to psychiatric service.
Alcohol consumption is a public health problem among the elderly and was one of the main risk factors pointed out by multiple studies (15,(17)(18)23,26,29) . Added to ubstance abuse (15) , this practice exerts an irrefutable influence on the morbidity and mortality of this group.These harmful habits are intensified by the vulnerability of the elderly facing aging changes, loss of friends and family, loneliness, social isolation and financial difficulties (43) .
Besides, this practice has become uneventful and a habit adopted among men throughout their patch in life.This behavior generates anguish and guilt after consumption; it also exacerbates depressive effects and increasing impulsivity, reinforcing the probability of suicide.The consumption of alcohol and other substances may be a decisive aspect in the constitution of a suicidal experience.Therefore, it should not be investigated in isolation, but as a clear social factor when discerning suicide (26,44) .
In the case surveys conducted (15,23,29) , significant associations were detected between the use of psychotropic drugs and suicide in the elderly.This fact is attributed to the frequent presence of psychiatric comorbidities in this population, which is an important risk factor, in addition to the use of psychotropic drugs to relieve somatic dysfunctions (15) .Also, studies indicate an association between marital status and suicide, where married individuals prevail (11,16,21,24,29) .This phenomenon can be explained by the fact that relationships of elderly couples are often instituted in youth, conserved for years and kept during all the transitions of life, until old age (45) .However, long-lasting relationships can refer to marital problems, considered an important determinant for suicide in the elderly (20,26) .These marital issues involve problems with sexuality, loss of control over the relationship and feeling of failure in exercising the male role according to the patriarchal culture, which predisposes such male individuals to suicidal outcomes (46) .
Another condition related to suicide in the elderly were social problems (11,23,26) and family members (12) ; it reflects the impoverishment of Suicide in the elderly: approach to social determinants of health in the Dahlgren and Whitehead model Carvalho ML, Costa APC, Monteiro CFS, Figueiredo MLF, Avelino FVSD, Rocha SS.
of primary relationships when it comes to routine dynamics, making intolerable the sphere of coexistence.Besides, social disconnection generates potentially negative repercussions on their mental health, and because they feel unsupported and mostly emotionally unsupported, the elderly release themselves from the link with life and intend to anticipate its end (47) .It is noteworthy that these family and social conflicts predispose to violence -another determinant associated with suicide, which, without mediation, interferes in personal relationships, nourishes traumas and raises suicidal ideas (12) .In this population, there is a close relationship between ideation, attempted and consummated suicide.Among the elderly, approximately two to four attempts are made before the suicide itself (48) .Thus, the presence of ideation or attempts intensifies the risk of self-inflicted deaths, a fact confirmed by some authors (11,18,22,24) ; it should thus be approached in greater depth, for it is directly related to consummated suicide and is a warning for possible later practice.
Also, regarding intermediate SDHs, another factor associated with suicide in the elderly was the absence of previous admissions to psychiatric service (21) , which exposes the neglect of the family or the elderly regarding the treatment of mental disorders -factors that may drive suicide -, apart from the difficulty of attending the services.The places should seek strategies that bring the elderly closer to programs aimed at a specific care for this public.This would surely solve the difficulties and promote prevention (11) .
With information on intermediate SDHs that have an impact on self-inflicted death in the elderly, we note the importance of redirecting policies that favor good development conditions for individuals, focusing on health conditions, life habits, social ties and networks of support.In addition to health care, psychosocial attention must be given to building community bases, which interact in the entire individual's process of development and influence the well-being of the elderly population.
Factors such as social inequality, low income, financial failure, and schooling influence the occurrence of suicide.A probable explanation is that social and economic conditions drive different material life models, with varying degrees of exposure to environmental risk factors and access to resources.This causes changes in behavioral and psychosocial aspects, such as perception of violence, feelings of deprivation, and stress.Thus, economic factors and low education level influence the individual's health, including his mental health (4) .
The mention of these aspects refers to a deficit in social support, which consists in the failure of governments to minimize stressful events related to the social environment.These stressors are usually basic sanitation, housing characteristics, and job opportunities (49) .
The identified economic, social and cultural macrodeterminants denote a context that includes stratification mechanisms, defining the individual socioeconomic hierarchy and weakening access to resources.This also means that individuals at a social disadvantage are subject to different risks.Through integrated intersectoral policies, measures that address the essential needs and rights of the population in the various contexts of State action will contribute to a healthy aging.

Study limitations
It is emphasized as a limitation of the present study the noninclusion of other databases, since other articles that were not indexed in the bases of choice could not be selected to constitute the sample.

Contributions to the fields of Nursing, Health, or Public policy
Deepening and recognizing the SDHs related to suicide in the elderly is fundamental both for the adherence of preventive measures, essentially of modifiable factors, and for the definition of health policies that provide resolutive care, with the valorization of the elderly in a perspective of integrality.The relevance of investing in risk reduction is emphasized; it is considered the network of social and cultural factors in which individuals are inserted.

CONCLUSION
It was observed that proximal determinants have more effects on the practice of suicide, especially mental disorders, with emphasis on depression.In addition, physical problems were also listed as predisposing factors to the occurrence of suicide.As for Intermediate determinants, they are mainly composed of modifiable factors and subject to interventions, since they are listed as alcohol use, substance abuse, family, marital, and social problems.Now as for Distal determinants, lower associations were found, and these SDHs' mechanisms of action were not recognized as presuming suicide in the elderly.

Chart 1 -
Stratification of the research question following the PICo strategy, Teresina, Piauí, Brazil, 2019

Suicide in the elderly: approach to social determinants of health in the Dahlgren and Whitehead model Carvalho
ML, Costa APC, Monteiro CFS, Figueiredo MLF, Avelino FVSD, Rocha SS.

in the elderly: approach to social determinants of health in the Dahlgren and Whitehead model
Carvalho ML, Costa APC, Monteiro CFS, Figueiredo MLF, Avelino FVSD, Rocha SS.