SciELO - Scientific Electronic Library Online

vol.17 suppl.2Health sector reform and social determinants of health: building up theoretical and methodological interconnections to approach complex global challengesRace and perceived racism, education, and hypertension among Brazilian civil servants: the Pró-Saúde Study author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Epidemiologia

Print version ISSN 1415-790X

Rev. bras. epidemiol. vol.17  supl.2 São Paulo  2014 

Original Articles

Social factors associated with mental disorders with risk situations in the primary health care

Bruno Lopes da Costa Drummond I  

Antônio Leite Alves Radicchi II  

Eliane Costa Dias Gontijo II  

IMunicipal Health of Belo Horizonte - Belo Horizonte (MG), Brazil

IIDepartment of Preventive and Social Medicine, School of Medicine, Universidade Federal de Minas Gerais - Belo Horizonte (MG), Brazil



To evaluate patients with mental disorders, with or without risk situations, treated at primary health care (PHC) units.


A cross-sectional study was performed in samples of 240 patients living in a region of high social vulnerability in Belo Horizonte. The response variable was mental disorders with risk situations (MD-WR). The explanatory variables were gender, age, marital status, literacy, education, employment, social benefits and per capita income. Instruments from Berkman and Syme (social network), Sherbourne and Stewart (social support), adapted for Brazil, were applied. Pearson's χ2 test and binary logistic regression were used for the adjusted analyzes.


The factors associated with MD-WR were being male (OR = 3.62; 95%CI 1.84 - 7.09); having "up to one confident relative" only (OR = 2.53; 95%CI 1.18 - 5.42); being "not able to return home" when away from their living area (OR = 3.49; 95%CI 1.40 - 8.71). The reduction in the affective dimension of the Medical Outcomes Study (MOS) scale increases the chance of MD-WR.


The availability and access to social and support networks are lower for patients with MD-WR and need to be strengthened to promote autonomy and citizenship among its users. We conclude that there is the need of public policies to increase the availability of social networking equipment and social support projects, encouraging the participation of families.

Key words: Social networking; Social support; Mental disorders; Primary health care; Life conditions; Social Determinants of Health


Severe mental illnesses, a priority for public policies on mental health, are, in most cases, characterized by exhibiting risk situations, which underline their clinical severity. Social factors are important in the evolution of cases and directly affect the daily lives of patients, hindering and even preventing them from carrying out daily activities. In this sense, social factors influence the approach to mental disorders, and it is necessary to evaluate them in order to optimize the care provided.

In scientific literature, severe cases receive different names: "common disorders that cause severe disability", "patients with severe mental illness", "severe cases of mental suffering", "major mental disorders", "severe mental disorders" or "chronical mental disorders". However, the most common are Severe Mental Illnesses (SMI) and Severe and Persistent Mental Illnesses (SPMI).

The most accepted definition of SMI is the one proposed by american group of the National Institute of Mental Health (1987), composed of a triad: diagnosis (nonorganic psychosis and personality disorder); duration (over two years); disability (moderate or severe, as measured by the Global Assessment of Functioning - GAF). However, there is no consensus and there is significant dispute as to the diagnoses to be included, as well as the degree of disability and especially the duration of the disease.

A review of literature by Schinnar et al.1, 1990, shows 17 definitions for SPMI, from 1972 to 1987, which leads to a wide variation in prevalence rates (4% to 88%) according to the criterion used. In 2000, Rugeri et al.2 found that SPMI definitions are inconsistent and non-operational. They concluded that, among patients with SPMI, 40% are not psychotic.

The specifications of the Spanish Association of Neuropsychiatry3, 2002, also point to the low consistency, in some studies, in applying the criteria for severe mental illnesses in practice. In 2006, Parabiaghi et al.4, comparing groups, concluded that "using the definition of SPMI seems to be more informative than the diagnosis of psychosis" to identify people with high levels of mental illness. The authors comment on recognizing SPMI as an interdiagnosis category, and not as a group. There are also British studies5 that include the diagnosis of severe neurosis in the definition, in order to operate the concept of SMI.

In Brazil, Cavalcanti et al.6 points that there is no consensus in the literature on the definition of SPMI. However, administrative rules from the Brazilian Ministry of Health classify SPMI patients as eligible for Psychosocial Care Centers (Centros de Atenção Psicossocial - CAPS).

According to Malmgren et al.7 in the International Journal of Law and Psychiatry, the term "severe mental illness" should be released from its decisive role in the legal medical process, due to the ambiguity of the concept and lack of scientific validity.

Such conceptual imprecision generates various definitions of SMI, often inconsistent and non-operational, as well as imprecision in the interdiagnosis category. These facts produce:

  • various classifications;

  • lack of uniformity among International Code of Diseases (ICD) included in the SMI group in official documents of public mental health policies;

  • non inclusion in the planning of severe neurotic illnesses, depending on the definition;

  • limitations for population-based studies due to the absence of questionnaires that define the response variable SMI.

The importance of studying this population resides in its prolonged evolution, sometimes for a lifetime, which causes constant periods of incapacity to work and even to operate socially, reflecting on family dynamics. The quality of life of the individual affected is limited, which can result in isolation and social exclusion. To maintain the treatment of these individuals, intersectoral actions with a plurality of health professionals is necessary.

In Brazil, the Ministry of Health8 estimates that 5 million people require continuous mental health care, in terms of onsets such as psychosis, severe mood illnesses, severe neuroses and severe mental retardation. The importance of this care is ratified by the National Health Agency (NHA)9, the regulating body of Supplemental Health in Brazil. ANS does not include the diagnosis of severe neurosis in the SMI group; only the functional psychoses, such as schizophrenia and disorders schizotypal, delirious and severe mood disorders.

SMI produce a strong economic impact on societies, including the cost of care and services due to prolonged psychological distress, representing a significant number of disability-adjusted life years (DALYs). The neuropsychiatric disability component of the global burden of disease (GBD) accounts for almost a third of all years lived with disability, according to the World Health Organization (WHO)10.

Given this conceptual gap, this study was based on risk situations and priorities of mental health policies to assess the severe cases. The intention is not to define SMI by risk situations, but to seek resources to aggregate the severe cases in primary health care (PHC). Risk situations, as well as indicators of vulnerability and severity of the case, allow the identification of those in need of rehabilitation, reintegration and social inclusion, as counterpoint to valuing dimensions such as "diagnosis", "disease duration" and "degree of disability".

Such examples of priorities of mental health policies and risk situations are articulated in the following documents: "Oficina de Trabalho para Discussão do Plano Nacional de Inclusão das Ações de Saúde Mental na Atenção Básica"11 and "Saúde Mental e Atenção Básica: O Vínculo e o Diálogo Necessários"12, such as, among others, "patients discharged from psychiatric hospitals; population living on the streets; patients who attempted suicide; patients treated at CAPS".

Another priority, the "insane offender", treated at the Psychosocial Care Center (CAPS) and at Primary Health Care units in partnership with the attention program for legal patients (PAI-PJ/MG), is described in the following documents: "Reforma Psiquiátrica e Política de Saúde Mental no Brasil" and "Saúde Mental no SUS: Acesso ao Tratamento e Mudança no Modelo de Atenção"13. In the latter, it follows the guidance from "Saúde Mental: Nova Concepção, Nova Esperança", by WHO. In PHC, there is also the express prioritizing for the treatment of epilepsy, due to its potential for risk situations and the stigmatization of the patient.

Although emergency care is the responsibility of services such as CAPS, the maintenance and stabilization of patients still in crisis (who could potentially be admitted) is carried out as a partnership with PHC, a fact determined by the document "Política de Saúde Mental de Belo Horizonte"14, which prioritizes "patients discharged from psychiatric hospitals and those who, in the recent past, could potentially be admitted" as their target public. Thus, the study was guided by the priorities and risk situations envisaged in official public mental health policy documents in Brazil, separated in a table after the temporality of the events was included for precision (Chart 1).

Chart 1  Description of risk situations, according to guidelines of mental health policies in Brazil, 2010. Source: referenced and adapted from official mental health policies documents.  

Using these risk situations as a reference, it was possible to identify two groups of patients routinely treated in Basic Health Units: 1) Mental Disorders with Risk Situations (MD-WR): presence of one or more risk situations, and 2) Mental Disorders with No Risk Situations (MD-NR): no risk situations.

The present study sought to compare them in relation to access to the social network and support among the Social Determinants of Health (SDH)15, prepared by the Brazilian Commission on Social Determinants of Health. The commission chose the Dahlgren and Whitehead model in its implications: proximal (related to lifestyle), intermediate (related to living, working, and social network and support conditions) and distal (related to the socioeconomic, cultural and environmental macrostructure).

The importance of social networks and support is such that, consistent with previous study16, it confirmed that people with no social ties were more likely to die during follow-up than those with greater ties, and that the association between social ties and mortality was independent of the state of self-perceived health and practices such as smoking, drinking and obesity.

There is also another study17 on the construct validity of the social support scale of the Medical Outcomes Study (MOS), conducted in Rio de Janeiro, 2005, confirming that less lonely individuals, with more frequent reporting of social group activities and with no evidence of Common Mental Disorder (CMD), would have a greater perception of social support. In this study, whose concepts were used here, a social network is defined as "a group of people with whom an individual maintains contact or some form of social bond that may or may not offer help in various situations throughout life", and social support as "resources made available by other people in situations of need and can be measured by the individual's perception of the degree to which interpersonal relationships correspond to certain functions (e.g., emotional, material and affective support)".

Finally, the available studies suggest an inverse relationship between the presence of social network and support and the presence of and/or increased psychiatric symptoms. The present study hypothesizes that the availability of social factors (mainly network and support) is smaller for MD-WR than for MD-NR, therefore seeking to assess its importance in patients with MD-WR treated by PHC.


This is a comparative study of social factors in patients with mental disorders who present or not risk situations, according to the priorities for mental health policies.

A total of 240 patients aged over 15 years old participated in this study, sampled according to the following criteria: having received treatment by mental health services, from July 2006 to December 2008, in one of the four Basic Health Units (BHU) in the region called Aglomerado da Serra, in Belo Horizonte; be residing in a reference area of the Health Units with a health vulnerability index (HVI) considered of "very high risk"18; seeking mental health services spontaneously or by referral by Family Health Teams (FHT), psychiatric hospitals, CAPS and the judiciary.

The sample size calculation was based on the prevalence of Common Mental Disorders: approximately 30% in the group with high social support and 50% in the group with low social support. The following parameters were used: significance level of 95%; sample's detection power of 80%; ratio of two (2) MD-WR per MD-NR and expected Odds ratio (between 2.0 and 2.2). Considering the lowest sample value calculated (n = 228) and a percentage of 5% losses, the sample was composed of 240 people.

The stratified sample procedure was applied, proportional to the representation of each BHU in the total population served in the four BHUs in the period from July 1st, 2006 to December 31st, 2008. Upon calculation of the sample, patients with MD-WR and MD-NR were selected in each BHU. Participants were interviewed by six community health agents (CHA), professionals who worked in the units and were indicated by their respective managers. Interviewers were trained by the first author for two weeks, by reading training materials and with the application of simulation tools.

Information on social support was collected by the instrument developed by Sherbourne et al., for the MOS, which is a Likert five-point scale consisting of 19 items and five dimensions of support: material, emotional, positive social interaction, information, affective. For information on the social network, the instrument prepared by Berkman et al. was used, consisting of five items: questions about the number of relatives and friends, participation in athletics, community activities, volunteer work. Positive responses refer to the five-point scales.

Both instruments were evaluated and validated in Brazil by the Pro-Health Research Group (RJ): Social Determinants of Health and Disease19.

In the interviews, sociodemographic variables such as gender, age, marital status, literacy, schooling, employment, social benefits and income per capita, and variables on social determinants, such as lifestyle (prevalent everyday activity, domestic and food habits, manual activities, use of tobacco and alcohol) and living conditions (ability to move in and out of the place of residence and security) were collected. To ensure that there was no mixing of the two groups in the period between the identification of the presence or absence of risk situations and the time of the interviews, all subjects classified as MD-NR were reassessed.

For categorical variables, frequency distributions were made and, for continuous variables, central tendency and variability were measured.

Along the affective and positive social interaction dimensions, the emotional and information dimensions were also evaluated, for being considered in the validation of the scale in Brazil. The scores were standardized according to the number of questions that are part of each dimension; the sum of scores obtained under each dimension is divided by the maximum possible score in the same dimension. The result of the ratio (points scored/maximum score of the scale) was multiplied by 100.

For comparison between groups, univariate analysis was performed initially, and for categorical variables, the χ2 test or Fisher's exact test were performed, as well as the estimated odds ratio, with 95%CI. For continuous variables, a nonparametric Mann-Whitney test was carried out. The control of possible confounding factors was performed by binary logistic regression. For entry of predictor variables in the logistic model, the p-value of 0.20 was used, and, for permanence, a 5% significance level. The model adjustment was assessed using the Hosmer-Lemeshow test. In all analyzes, the 5% significance level was considered. The software used was Statistical Package for the Social Sciences (SPSS) 15.0.

This study was approved by the Research Ethics Committee of the Municipal Department of Health of Belo Horizonte and the Ethics Committee of Universidade Federal de Minas Gerais (UFMG), under the following protocol numbers, respectively: no. 0227.0.203.203-09A and ETIC 0227.0.410.203-09. Participants signed the free and informed consent form after receiving information on the study.


Among the 240 people, there were 11 refusals, and 229 were interviewed, 76.4% of them being women. The mean age was 45.6 years and median was 45, with no differences between groups. Most individuals have no partner (55%). The rate of illiteracy is 43.3%, with a median of 4 years of study. The average time of employment was 8.3 years, and 39.8% do not receive any social benefit and 14.3% were given disability retirement. The average income per capita was of 0.4 to three minimum wages/month, with a minimum of R$ 0,00 and maximum of R$ 1,395.00 (at an exchange rate of 1 dollar = 2.18 reais at the time of the study). There was a greater percentage of men among patients with MD-WR (37%) than in those with MD-NR (17%), and the opposite occurs with women.

The sociodemographic variables indicate that being male increases three times the chance of MD-WR, and that people with MD-WR are around twice as likely to be illiterate and retired for disability (Table 1).

Table 1 Sociodemographic variables for mental disorders with risk situations and mental disorders with no risk situations. Belo Horizonte, Minas Gerais, Brazil, 2008 - 2010. 

Variable MD-NR MD-WR p-value OR [95%CI]
n (%) n (%)
Male 25 (16.7) 29 (36.7) 0.001* 2.90 [1.55 – 5.43]
Female 125 (83.3) 50 (63.3) 1.00
Can read, write and do math 84 (63.2) 34 (45.3) 0.013* 1.00
Can do one or two of the above skills 49 (36.8) 41 (54.7) 2.07 [1.16 – 3.67]
Social benefits
None 51 (37.2) 33 (44.6) 0.297* 0.74 [0.41 – 1.31]
Disability retirement 14 (10.7) 15 (20.8) 0.018* 2.27 [1.03 – 4.99]

*Pearson's χ2 test. MD-NR: mental disorders with no risk situation; MD-WR: mental disorders with risk situation; OR: Odds Ratio; 95%CI: confidence interval of 95%.

In the lifestyle determinant, some variables such as the number of people per household, domestic and food habits and manual abilities, showed no significant associations. However, for MD-WR participants, there are three significant (p = 0.001) categories in variable "prevalent everyday activity": lower chance of "doing some activity outdoors"; lower chance of "working with a rap sheet"; greater chance of "wandering around in the streets, doing nothing". There is also borderline significance value for use of alcoholic beverages.

In living conditions, variables related to security in residing, living and transiting in the neighborhood were not significantly associated. However, individuals with MD-WR have a greater chance of "being unable to return home alone when they leave the neighborhood", and there is still a borderline significance for the variable "being unable to go where needed alone when they leave the neighborhood" (Table 2).

Table 2 Lifestyle and living conditions for mental disorders with risk situation and mental disorders with no risk situation. Belo Horizonte, Minas Gerais, Brazil, 2008 - 2010. 

Variable MD-NR MD-WR p-value OR [95%CI]
n (%) n (%)
Relationship with alcohol and tobacco
Smokes or has smoked 61 (40.7) 41 (51.9) 0.104* 1.57 [0.91 – 2.72]
Uses or has used alcohol 44 (29.5) 33 (41.8) 0.063* 1.71 [0.97 – 3.02]
Is able to go where needed when leaves the neighborhood
Yes 128 (85.9) 60 (75.9) 0.060* 1.00
No 21 (14.1) 19 (24.1) 1.93 [0.97 – 3.86]
Is able to go back home when leaves the neighborhood
Yes 139 (92.7) 65 (82.3) 0.017* 1.00
No 11 (7.3) 14 (17.7) 2.72 [1.17 – 6.32]

*Pearson's χ2 test. MD-NR: mental disorders with no risk situation; MD-WR: mental disorders with risk situation; OR: Odds Ratio; 95%CI: confidence interval of 95%.

In social network, variables analyzed, such as confidant friends, participation in sports or artistic activities, meetings, volunteer work and knowledge of the local social networking, showed no significant associations. However, the variable "confidant relatives" was included in the multivariate model for presenting p = 0.202.

In social support, the perception of improved health and family participation in treatment showed no significant association. However, of the five dimensions assessed by the MOS scale, individuals with MD-WR show statistically significant reduction in the affective dimension and borderline values in the affective + positive social interaction dimension (Table 3).

Table 3 Social support for mental disorders with risk situation and mental disorders with no risk situation. Belo Horizonte, Minas Gerais, Brazil, 2008 - 2010. 

Variable MD-NR MD-WR p-value
Dimensions* Mean Median Mean Median
1st – affective 84.1 93.3 78.2 80.0 0.024**
2nd – emotional 67.6 65.0 67.7 65.0 0.940**
3rd – information 72.4 75.0 68.7 65.0 0.103**
4th – positive social interaction 58.0 55.0 54.1 50.0 0.244**
5th – material 79.3 85.0 79.9 85.0 0.612**
Affective + positive social interaction 69.2 65.7 64.4 62.9 0.066**
Emotional + information 70.0 67.5 68.2 67.5 0.421**

*Likert scale

**Mann-Whitney's test; MD-NR: mental disorders with no risk situation; MD-WR: mental disorders with risk situation.

The analysis confirms that being male increases nearly four times the chance of having MD-WR. Individuals with MD-WR showed over thrice the chance of "being unable to return home alone when they leave the neighborhood" and an important deficiency in social network, with a greater chance of having only up to one confidant relative", and in social support, with a greater chance of reduced scores in the affective dimension (Table 4).

Table 4 Multivariate model considering the condition as the response variable (mental disorders with risk situation/mental disorders with no risk situation). Belo Horizonte, Minas Gerais, Brazil, 2008 - 2010. 

Variable p-value OR 95%CI
Female (reference) < 0.001 1.00
Male 3.62 [1.84 – 7.09]
Leaves the neighborhood and is able to go back home
Yes (reference) 1.00
No 0.007 3.49 [1.40 – 8.71]
Confidant relatives
Up to one 0.017 2.53 [1.18 –; 5.42]
Two 0.259 1.69 [0.68 – 4.23]
Three or more (reference) 1.00
Affective dimension (reduction of one unit in the score) 0.042 1.015 [1.001 – 1.030]

Adjustment of the model by Hosmer & Lemeshow (p-value = 0.680). OR: Odds Ratio; 95%CI: confidence interval of 95%.

The reduction in the affective dimension in the MOS scale is associated with an increased chance of it being a MD-WR case. The reduction of one unit in the MOS scale increases such chance in 1.015 times, ranging between 1.001 and 1.030. As this relationship is multiplicative, it can be stated that the reduction of 10 units in the affective dimension of the MOS scale is associated with a 10.15 times greater chance of it being a MD-WR case.


There are few studies on the treatment of serious mental illnesses in PHC, especially those that include, in addition to strategies to mitigate the disability caused by the disease, the conditioning social factors of the disorders. Moreover, the lack of consensus on the definition of severe mental illnesses raises difficulties for the conduction of further studies that contribute both to the planning of care and to provide guarantees of rights and citizenship.

The effort of the present study was to shed light on the importance of social factors in cases of mental disorders treated in primary care. The lack of precise diagnostic criteria for defining SMI could be a limitation. However, the strategy of using the definitions found in official documents that define public policies for the area, and the use of risk situations, highlights the clinical severity of these cases.

It has been demonstrated in the literature that women have a greater association with mental illnesses20, and that there is a higher prevalence of men with SMI (mainly schizophrenia), both in the primary health care21 and in admissions to mental hospitals22. This study demonstrates that there is a significant association between being male and presenting MD-WR in primary health care.

Illiteracy (54.7%) and disability retirement have a significant association with MD-WR. It is possible that this finding is due to the presence of risk situations, such as hospitalization, legal issues, crises that repeatedly end up limiting learning and produce disability. A review of literature study states that the income received by schizophrenic patients is shown to be associated with a better quality of life23. Disability retirement, in this study, is a valuable source of income compared with the low coverage of other social benefits, such as scarcity in invalidity aid (1.4%) and in the "De volta para casa" program (0%) in the MD-WR group.

Regarding lifestyle, the use of alcoholic beverages can aggravate a risk situation due to the metabolic syndrome24, as well as increased doses of psychotropic drugs to which patients with MD-WR are submitted. In living conditions, there is significant association with "being unable to return home alone when they leave the neighborhood". This finding is important for a more comprehensive care in PHC, as it is known that there is a need for access to various services outside the space where these people live. Thus, we need to ensure the "coming and going" of these individuals, with transportation and therapeutic companions, for their access to the social networks and equipment necessary for treatment strategies.

This study also demonstrates the fragility of the social network, with a significant association between presenting a case of MD-WR and having only up to one confidant relative (62%). This fact attested by a study in São Paulo, in 2005, that highlighted the fragility of social networks of people with SMI and the importance of including strategies for the activation of social networks in the context of welfare projects25. These data provide information on the scarce presence of the family, often limited, in helping with these severe cases that require care in their daily lives.

With regard to social support, this study confirms that the reduction in the affective dimension of the MOS scale increases the chance of MD-WR. Therefore, there is, in respondents with MD-WR, the perception of a decrease in the affective dimension in their interaction with other people. Social support is "resources made available by other people", therefore essential to operate the social network, which is the "group of people with whom an individual maintains some degree of contact". With the fragility of the affective dimension of social support, subjects with MD-WR may find themselves in a situation of social isolation. This fact points to the need for more training and involvement of people living who live with individuals and of people leading the treatment of these cases. Similar averages, in this study, those found in MOS26 (on the scale of social support) for affective and emotional/information dimensions probably are due to the presence of chronic diseases in the two groups.

Thus, it was found that there is greater chance of men with mental disorders to present risk situations, and therefore greater severity. Limitations were observed in circulation outside of the space where they live, as well as a restriction of the trust relationship established with only one relative and reduced expression of affective dimensions in their personal relationships. This situation confirms that the availability of the social network and support is lower for patients with MD-WR than for MD-NR, and suggests the importance of social ties in patients with mental disorders.


The management of patients with mental disorders has historically been a major challenge, particularly in primary care, which led, in contemporary times, the effort to establish public policies that address the technical quality and humanization of such care.

Issues of social disintegration are not necessarily the direct causes of mental disorder. However, the reduced availability of access to social network and support contribute to complicate the context in which the subject lives, interfering directly in the management of their treatment.

The treatment plan for these patients involves their inclusion in social spaces, the identification of resources needed to establish intersectoral action on social network and support (family, legal, welfare, safety, artistic arts, sports), and multidisciplinary care shared with family health teams in clinics supervised by mental health teams. This integral approach should contribute to the reduction of the spending of public and family money, as well as to reduce the loss of years of disability, allowing the qualification of care and better quality of life.

The attention to mental health at the institutional end reveals that symptoms structured over the lifetime of a subject, fueled by social deprivation and demonstrated on a threshold event, make up the clinical complexity and highlights the severity of mental suffering.

The efforts of public policies directed at these risk situations should be doubled in order to enhance and make the care provided in primary care more efficient, minimizing social exclusion and promoting autonomy and citizenship to individuals with mental disorders.


. Schinnar AP, Rothbard AB, Kanter R, Jung YS. An empirical literature review of definitions of severe and persistent mental illness. Am J Psychiatry 1990; 147(12): 1602-8. [ Links ]

. Ruggeri M, Leese M, Thornicroft G, Bisoffi G, Tansella M. Definition and prevalence of severe and persistent mental illness. Br J Psychiatry 2000; 177: 149-55. [ Links ]

. Asociación Española de Neuropsiquiatría, Rehabilitación Psicosocial del Trastorno Mental Severo - Situación Actual y Recomendaciones, 2002, Cuadernos Técnicos 6. p. 5-8. [ Links ]

. Parabiaghi A, Bonetto C, Ruggeri M, Lasalvia A, Leese M. Severe and persistent mental illness: a useful definition for prioritizing community - based mental health service interventions. Soc Psychiatry and Psychiatric Epidemiology 2006; 41(6): 457-63. [ Links ]

. Slade M, Powell R, Strathdee G. Current Approaches to Identifying the Severely Mentally Ill. Soc Psychiatry Psychiatry Epidemiology 1997; 32: 177-84. [ Links ]

. Cavalcanti MT, Dahl CM, Carvalho MCA, Valencia E. Critérios de admissão e continuidade de cuidados em centros de atenção psicossocial- CAPS, Rio de Janeiro, RJ. Rev Saúde Pública 2009; 43(Supl 1): 23-8. [ Links ]

. Malmgren H, Radovic S, Thóren H, Haglund B. A philosophical view on concepts in psychiatry. International Journal of Law and Psychiatry 2010; 33(2): 66-72. [ Links ]

. Ministério da Saúde do Brasil. Reforma psiquiátrica e Política de Saúde Mental no Brasil; Conferência Regional de Reforma dos Serviços de Saúde Mental - 15 anos depois de Caracas. Nov. 2005. p. 42-44. Disponível em: . (Acessado em 13 de setembro de 2010). [ Links ]

. Ribeiro, JM. A Agência Nacional de Saúde Suplementar e as Políticas de Saúde Mental Direcionadas para Portadores de Enfermidades Mentais Severas. Documento Técnico da ANS, RJ, 2003. Disponível em: (Acessado em 16 de maio de 2012). [ Links ]

. World Health Organization. The World Health Report, 2001. Mental Health: New Understanding, New Hope, 2002, 1ª ed. p. 58-68. Disponível em: Acessado em 13 de setembro de 2010). [ Links ]

. Ministério da Saúde. Oficina de Trabalho para Discussão do Plano Nacional de Inclusão das Ações de Saúde Mental na Atenção Básica, 2001. Disponível em: (Acessado em 13 de setembro de 2010). [ Links ]

. Ministério da Saúde. Secretaria de Atenção à Saúde. Departamento de Ações Programáticas Estratégicas. Departamento de Atenção Básica. Coordenação Geral da Atenção Básica, Saúde Mental e Atenção Básica. O Vínculo e o Diálogo Necessários, 2003. Disponível em: (Acessado em 13 de setembro de 2010). [ Links ]

. Ministério da Saúde do Brasil. Secretaria de Atenção a Saúde. Relatório de Gestão, 2003-2006. Saúde Mental no SUS: acesso ao tratamento e mudança no modelo de atenção, 2007. Disponível em: (Acessado em 13 de setembro de 2010). [ Links ]

. Ministério da Saúde. Secretaria de Atenção a Saúde. Política Nacional de Humanização. Política de Saúde Mental de Belo Horizonte/MG, 2006. Disponível em: (Acessado em 13 de setembro de 2010). [ Links ]

. Comissão Nacional sobre Determinantes Sociais da Saúde. As Causas Sociais das Iniqüidades em Saúde no Brasil. Rio de Janeiro: Fiocruz; 2008. p.20-21. [ Links ]

. Berkman LF, Syme SL. Social network, host resistance, and mortality: a nine-year follow-up study of Alameda County residents. Am Journal of Epidemiology 1979; 109(2): 186-204. [ Links ]

. Griep RH, Chor D, Faerstein E, Werneck GL, Lopes CS. Validade de construto de escala de apoio social do Medical Outcomes Study adaptada para o português no Estudo Pró-Saúde. Cad Saúde Pública 2005; 21(3): 703-14. [ Links ]

. Escola de Saúde Pública do Estado de Minas Gerais. Oficinas de Qualificação da Atenção Primária à Saúde em Belo Horizonte: territorialização e diagnóstico local. Belo Horizonte: Escola de Saúde Pública do Estado de Minas Gerais; 2010. p. 16. [ Links ]

. Chor D, Griep RH, Lopes CS, Faerstein E. Medidas de rede e apoio social no Estudo Pró-Saúde: pré-testes e estudo piloto. Cad Saúde Pública 2001; 17(4): 887-96. [ Links ]

. Araújo TM, Pinho OS, Almeida MMG. Prevalência de transtornos mentais comuns em mulheres e sua relação com as características socio-demográficas e o trabalho doméstico. Rev Bras Saúde Matern Infant 2005; 5(3): 337-48. [ Links ]

. Santos DN, Modesto TN. Saúde Mental na Atenção Básica. Rev Baiana de Saúde Pública 2007; 31(1): 19-24. [ Links ]

. Lyra da Silva JP, Coutinho ESF, Amarante PD. Perfil demográfico e sócio-econômico da população de internos dos hospitais psiquiátricos da cidade do Rio de Janeiro. Cad Saúde Pública 1999; 15(3): 505-11. [ Links ]

. Araújo de Souza L, Coutinho ESF. Fatores associados à qualidade de vida de pacientes com esquizofrenia. Rev Bras Psiquiatr 2006; 28(1): 50-8. [ Links ]

. Rocha FL, Teixeira PJR. Associação entre síndrome metabólica e transtornos mentais Rev Psiq Clín 2007; 34(1): 28-38. [ Links ]

. Gulinelli A, Ratto RLC, Menezes PR. O ajustamento social de pessoas com transtornos mentais graves em São Paulo, Brasil. Rev Bras Psiquiatr, 2005; 27 (4): 309-14. [ Links ]

. Sherbourne CD, Stewart AL. The MOS Social support survey. Social Science and Medicine 1991; 32(6): 705-14. [ Links ]

Financing source: none

Received: May 23, 2012; Revised: December 06, 2012; Accepted: June 12, 2013

Corresponding author: Bruno Lopes da Costa Drummond. Municipal Health of Belo Horizonte. Rua Domingos Vieira, 273, sala 905, Santa Efigênia, CEP: 30150-240, Belo Horizonte, MG, Brasil. E-mail:

Conflict of interests: nothing to declare

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.