Equity in the coverage of health expenses by the Brazilian Unified Health System for people with signs of common mental disorders in the city of São Paulo

ABSTRACT:

Introduction:

The objective of this investigation was to check equity in coverage/complementation of health expenses by the Brazilian Unified Health System (SUS) for people with signs of common mental disorders, in 2008, in the city of São Paulo, Brazil.

Methods:

Cross-sectional study with data from the 2008 São Paulo Health Survey. Individuals aged 16 or older and with signs of common mental disorders were selected and evaluated with the Self-Reporting Questionnaire (SRQ-20). The following aspects were analyzed: demand for SUS, coverage of health expenses by SUS, and health expenses in the previous month, withal according to sociodemographic and health conditions.

Results:

The search for SUS services was lower among white people with higher per capita incomes, in a stable union, and with higher education degrees. Coverage by SUS was lower for people aged between 45 and 59 years and aged 60 years or older, with higher per capita incomes, with high or technical school, and college degrees. Subjects who spent more on family health were 60 years or older, white, with high per capita income, in a stable union, and with a college degree.

Conclusions:

Among people with signs of common mental disorders, SUS mainly services and covers the expenses of the poorer population with lower educational level; therefore coverage is unequal and favors who needs it the most. However, considering age, inequity became explicit, since SUS was shown to offer wider coverage of health expenses to the youngest population.

Keywords:
Health expenditures; Health equity; Health status disparities; Mental disorders; Cross-sectional studies

RESUMO:

Introdução:

O objetivo deste trabalho foi verificar a equidade na cobertura/complementação dos gastos com saúde pelo Sistema Único de Saúde (SUS) de pessoas com indicativos de transtornos mentais comuns (TMC), no ano de 2008, no município de São Paulo.

Métodos:

Foi realizado um estudo de corte transversal a partir dos dados do Inquérito de Saúde no Município de São Paulo de 2008 (ISA-Capital 2008). Os sujeitos incluídos na pesquisa tinham 16 anos de idade ou mais e indicativos de TMC, avaliados por meio do instrumento Self-Reporting Questionnaire (SRQ-20). Foram analisados: a procura pelo SUS, a cobertura dos gastos com saúde pelo SUS e os gastos com saúde no último mês, correlacionando-os com aspectos sociodemográficos e de condições de saúde.

Resultados:

A procura pelo SUS foi menor entre as pessoas de cor branca, renda per capita elevada, com união estável e Ensino Superior. A cobertura pelo SUS foi menor entre as faixas etárias de 45 a 59 anos e de 60 anos ou mais, com renda per capita elevada, Ensino Médio ou Técnico e Ensino Superior. As pessoas que gastaram mais com a saúde da família foram aquelas com idade igual ou superior a 60 anos, de cor branca, renda per capita mais elevada, com união estável e Ensino Superior.

Conclusões:

Foi observado que o SUS atende e cobre os gastos majoritariamente daqueles com renda per capita e escolaridade mais baixas, denotando uma cobertura desigual que favorece os mais necessitados. Porém, considerando o fator idade, ficou explícita uma situação de iniquidade, pois foi constatada maior cobertura dos gastos em saúde pelo SUS para a população mais jovem.

Palavras-chave:
Gastos em saúde; Equidade em saúde; Desigualdades em saúde; Transtornos mentais; Estudos transversais

INTRODUCTION

The Brazilian Constitution of 1988¹ acknowledges health as a citizen right and State’s duty. The creation of the Brazilian Unified Health System (SUS) kick-started the implementation of a network providing universal access and aimed to ensure full and equal health and well-being for the Brazilian population.

Souza² points out that equity is a step forward in the discussion about equality, since the concept of equity includes respect for the right to equality, but prioritizes those who need it the most. This principle brings about the notion of difference and comes about in the current period associated with minorities, opposing to classical legal thinking when it comes to common and undifferentiated citizenship, putting diversity as a human condition and proposing that differences be treated as guides for public policies.

Margaret Whitehead³ sums a parameter of justice to the term “equity”. For her, inequities are unnecessary and avoidable differences, also considered abusive and unfair. Judgment and inequality measurements will depend on the concept of social justice adopted by societies and on the moment when one thinks about this issue. In general, equity “presupposes unequal treatment for those who are disadvantaged, which paves the way for what is considered a kind of ‘positive discrimination’⁴”

According to Buss and Pellegrini Filho⁵, life and work conditions of individuals are related to their health situation, which means that social, economic, cultural, ethnic/racial, psychological and behavioral factors function as social determinants and influence the onset of health issues and risk factors. On that, Travassos⁶ also considers that inequalities in living conditions expose people differently to health and disease determinants, emphasizing that health equity is linked to service access and use, resource allocation, and reduction of inequalities resulting from social statuses in medical services. For Campos⁷, health equity is also related to service quality, that is, efficacy and problem-solving ability. Thus, the core of public policies aiming at health equity is to reduce or eliminate differences arising from factors considered avoidable and unfair, thus favoring more equitable opportunities.

In order to investigate equity, it is important to question which aspects of the concept one wants to focus on, as well as to select a specific and homogeneous group. It should therefore be noted that, in this research, the equity in health expenses coverage/complementation by SUS for people with indications of common mental disorders (CMD) was studied. This analysis was based on data from a population-based and cross-sectional study Health Survey conducted in the São Paulo Municipality (ISA-Capital 2008).

According to Silva and Menezes⁸, CMDs have been increasingly identified and researched by health professionals because of their high prevalence in the community. The term common mental disorder, coined by Goldberg and Huxley⁹, refers to a health situation that does not meet enough criteria for a formal diagnosis of any mental disorders, but whose symptoms, such as insomnia, fatigue, somatic complaints, forgetfulness, irritability, difficulty concentrating and others, cause significant functional incapacitation and brings psychosocial damages to individuals in addition to high social and economic costs.

According to Lima¹⁰, studies conducted by the World Health Organization (WHO) have pointed out that persistent physical symptoms without medical explanation may be associated with mental health and that there seems to be a current tendency for people with mental disorders to seek health care more often. Despite this, according to Viana and Andrade¹¹, both mild and severe disorders are associated with inefficiency and lack of adequate treatment. Therefore, it is an important condition from the point of view of public health, as shown by Lima¹⁰, Moraes Júnior¹² and Pinheiro¹³.

METHODOLOGY

This study used data from a broader research named ISA-Capital 2008, a population-based and cross-sectional study whose purpose was to monitor the living and health conditions of people, and the use of health services. The sample was composed of people aged 16 years or older who answered to Block M on Emotional Health of ISA-Capital 2008 and were assessed as presenting indications of CMDs. Thus, 580 people were surveyed in 2008.

The presence of CMD was defined by the Self-Reporting Questionnaire (SRQ-20) score, referring to Block M - Emotional Health. SRQ-20 is an instrument recommended by WHO for population-based studies on emotional health¹⁴.

The cut-off point for CMD in this study was 6 or more positive responses for men younger than 64 years old, and 8 or more positive responses for women younger than 64 years old. For people aged 65 or over, the cut-off point was 5 or more positive responses for both genders¹⁶.

The outcomes investigated in this project were:

1. search for SUS services;

2. expenses coverage/complementation by SUS;

3. health expenses in the month prior to survey.

Data obtained through the C (questions C08 and C12) and R (R01 to R11) blocks of ISA-Capital 2008 served this purpose.

Two sets of independent variables were considered:

1. socio-demographic characteristics: gender, age, marital status of household head, educational level of household head, per capita income, household head origins, household head occupation, skin color, religion;

2. health conditions: health problem in the 15 days prior to survey, physical disability, chronic diseases, depression/anxiety (self-reported diagnosis), and headache.

Stata 11 (StataCorp, College Station, United States) and the svy group controls were used to analyze data, the procedures of the latter for population surveys including the necessary weights for populational samples. In the first phase, association between the outcome of interest and sociodemographic variables and health conditions of people with indications of CMD was assessed. Results were based on the Pearson’s χ² test and on the Poisson regression method.

In the second phase, the variables showing positive association with the dependent variable at a level of significance lower than 0.20 were included in the multiple regression analysis (based on the Poisson regression method) between the outcome of interest and sociodemographic and health conditions of people with indications of CMD. The persistent variables associated with the dependent variable were maintained in the final model, with level of significance lower than 0.05. In this paper, only the results of the multiple regression are presented.

RESULTS

A prevalence of 21.3% (95%CI 19.2 - 23.6) of people with indications of CMD was found, according to the population sample weighting. Considering the people who searched for a health service, 66 went for SUS and 41 for other services (information of 12 individuals missing). In total, 105 people were provided with care and only two were not assisted. Among people who received care, 65 had their expenses supplemented or covered by SUS and 40 by health plans, direct payments, a company or other means (Figure 1).

Figure 1:
Chart representing the population studied: people with indications of common mental disorders who presented with morbidities in the 15 days prior to data collection and searched for a health service.

As per the Poisson multiple regression, the demand for SUS services was significantly lower among white people when compared to the category “other” (encompassing brown, black, yellow or indigenous people); the category per capita income “> 1-2.5” minimum wages presented less chances of searching for SUS when compared to the income range “≤ 0.5”; household heads in a stable union (including “married”) had less chances of searching the SUS when compared to the category “other” (single, separated, divorced or widowed); and the demand for SUS was also significantly lower among people with Higher Education when compared to those who never attended school (Table 1).

Table 1:
Frequency of people with signs indicating common mental disorders who reported morbidities in the 15 days prior to the interview and searched for the Unified Health System, according to sociodemographic characteristics and health conditions, in 2008, as per the results of Poisson multiple regression.

Variables	Total (n)	Proportion of subjects who went to SUS for assistance*	PR**	95%CI***	p-value
Ethnicity1
Other	38	0.78
White	68	0.44	0.71	0.54 - 0.93	0.02
Family per capita income²
≤ 0.5	33	0.87
> 0.5-1	27	0.66	0.76	0.57 - 1.03	0.07
> 1-2.5	26	0.61	0.74	0.56 - 0.99	0.04
> 2.5	21	0.22	0.71	0.27 - 1.86	0.48
Marital status of household head²
Other	46	0.65
Stable union	61	0.5	0.71	0.55 - 0.92	0.01
Educational level of household head 2
None	12	0.83
Incomplete/complete basic education	63	0.76	1.08	0.80 - 1.46	0.61
Incomplete/complete high school/technical school	22	0.43	0.7	0.41 - 1.21	0.2
Incomplete/complete higher education	10	0.03	0.06	< 0.01 - 0.72	0.03
Total	119	_	_	_	_

• SUS: Brazilian Unified Health System; *proportion considering the weight for the populational sample; PR**: prevalence ratio; IC95%***: 95% confidence interval; ¹information of 13 subjects missing; ²information of 12 subjects missing.

Regarding coverage/complementation of expenses by SUS, in the final model, significant associations were seen: age ranges 45-59 years and 60 years or more had less chances of having expenses covered/complemented by SUS when compared to the age group of 16 to 29 years; per capita income ranges “> 0.5 to 1” and “> 1 to 2.5” minimum wages, had lower expense coverage/supplementation by SUS when compared to the income range “≤ 0.5”; finally, “Incomplete/complete high/technical school” and “Incomplete/complete Higher Education” categories had lower probability of being covered by SUS when compared to subject who never attended school (Table 2).

Table 2:
Frequency of people with signs indicating common mental disorders who reported morbidities in the 15 days prior to the interview and whose expenses were complemented or covered by the Unified Health System, according to sociodemographic characteristics and health conditions, in 2008, as per the results of Poisson multiple regression.

Variables	Total (n)	Proportion of subjects who had expenses covered/complemented by SUS*	PR**	95%CI***	p-value
Age1
16-29	15	0.85
30-44	18	0.62	0.6	0.31 - 1.03	0.06
45-59	10	0.43	0.39	0.20 - 0.78	< 0.01
60 or older	60	0.51	0.41	0.23 - 0.74	< 0.01
Family per capita income
≤ 0.5	32	0.87
> 0.5-1	27	0.73	0.74	0.56 - 0.98	0.04
> 1-2.5	25	0.56	0.46	0.29 - 0.73	< 0.01
> 2.5	21	0.34	0.55	0.24 - 1.26	0.15
Educational level of household head
None	12	0.83
Incomplete/complete basic education	62	0.75	0.83	0.54 - 1.28	0.39
Incomplete/complete high school/technical school	21	0.45	0.45	0.28 - 0.73	< 0.01
Incomplete/complete higher education	10	0.28	0.36	0.15 - 0.91	0.03
Total	105	_	_	_	_

• SUS: Brazilian Unified Health System; *proportion considering the weight for the populational sample; PR**: prevalence ratio; IC95%***: 95% confidence interval; ¹information of two subjects missing.

As to health expenses in the month prior to study, among people with CMD, the variables that remaned significantly associated with the outcome “more than R$ 100 spent on family health”, as per the Poisson multiple regression, were: age range above or equal to 60 years, spending more on health than the age group of 16 to 29 years; white-skinned people, also spending more with health; per capita income groups “> 1-2.5” and “> 2.5” minimum wages, spending more with health when compared to the range “≤ 0.5”; people in a stable union, spending more with health; and people with higher education degrees, also spending more on health than those who never attended school (Table 3).

Table 3:
Frequency of people with signs indicating common mental disorders who reported spending more than 100 reais with family health in the month prior to the interview, according to sociodemographic characteristics and health conditions, in 2008, as per the results of Poisson multiple regression.

Variables	Total (n)	Proportion of subjects who spent more than R$ 100 on healthcare*	PR**	95%CI***	p-value
Age1
16-29	112	0.42
30-44	103	0.51	1.13	0.86 - 1.48	0.37
45-59	70	0.48	1.15	0.87 - 1.52	0.31
60 or older	261	0.59	1.66	1.25 - 2.19	< 0.01
Ethnicity2
Other	216	0.37
White	329	0.58	1.29	1.03 - 1.61	0.02
Family per capita income²
≤ 0.5	161	0.29
> 0.5-1	143	0.4	1.29	0.84 - 1.96	0.24
> 1-2.5	145	0.52	1.61	1.10 - 2.36	0.02
> 2.5	99	0.75	1.95	1.24 - 3.05	< 0.01
Marital status of household head²
Other	227	0.36
Stable union	321	0.57	1.47	1.14 - 1.88	< 0.01
Escolaridade do chefe de família3
Nunca frequentou	49	0.23
Fundamental Incompleto/Completo	346	0.41	1.64	0.92 - 2.91	0.09
Ensino Médio/Técnico Incompleto/Completo	105	0.52	1.76	0.96 - 3.22	0.06
Superior Incompleto/Completo	48	0.82	2.26	1.27 - 4.03	< 0.01
Total	580	_	_	_	_

• *proportion considering the weight for the populational sample; PR**: prevalence ratio; IC95%***: 95% confidence interval; ¹information of 34 subjects missing; ²information of 35 subjects missing; ³information of 32 subjects missing.

DISCUSSION

In a study on the sociodemographic profile of users and pattern of use within SUS, Silva et al.¹⁷ found that, in 2008, the system accounted for 56.7% (95%CI 55.9% - 57.5%) of health care assistance provided in the country. These findings validate Silva et al.’s¹⁷: there were significant differences as to users’ ethnicity (higher demand for SUS among non-white people), income and educational level (the higher the income and educational level, the lower the search for SUS). Regarding gender, however, no significant differences were found. When it comes to marital status of household head, an explanation for married people seeking less often for SUS may be the possession of health plans, which reduces demand for public services, as pointed out by Silva et al.¹⁷,

The matter of health expenses coverage/complementation by SUS for people with indications of CMD is summed up by patients aged 45 to 59 years and 60 years or older being less covered when compared to the 16 to 29 year-old range. It is important to note that the demand for SUS was proportionally similar between all age groups, showing that the younger population that sought the private system subsequently had access to services offered by SUS. It can be thus verified that there was a double coverage and complementary offer by the public system for this age range, not contemplating the most advanced-age groups.

According to Bós and Bós,¹⁸ the public health system in Brazil is typically geared towards younger groups, especially infants and children, and with the growth of the elderly population, these individuals are expected to be more and more marginalized in care, if no changes that effectively involve this new demographic and epidemiological profile are made.

Silva et al.¹⁷ point out that SUS “accounts for the health of approximately 190 million individuals, while the health insurance segment takes 49.2 million, representing a doubled coverage for 25.9% of the population.” It is not unusual for people who have health insurance to use SUS in a complementary way for certain procedures (not all high-cost exams or medications are fully covered and many have their offer restricted by health insurance plans).¹⁹ In this scenario, expenses with prevalent middle-class demands increase, forcing the SUS to limit the expenses with procedures that could be used to reduce the morbidity and mortality of the most needed portion of population¹⁹.

Like the studies by Louvison et al.²⁰ and Bós and Bos¹⁸, our study focused on the population with indications of CMD and showed educational level and family income to be inversely proportional to coverage by SUS, that is, people with higher incomes are usually insured by private health plans and/or have access to direct payment possibilities. Thus, among the people who sought assistance at health services in the face of morbidities, we can say that there was inequality in coverage by the public network favoring users with lower per capita income, but it is important to note that the speed of care provided, ease of access or quality of service provided were not assessed in this matter. According to Granja et al.²¹, “in order to place individuals born under unequal situations under the same starting conditions, it may be necessary to favor the poorest to the detriment of the richest, with an unequal distribution.”

As to expenses with health in the month previous to survey by people with indications of CMD, families with members aged 60 years or more were more likely to have spent more than R$ 100 (prevalence = 59 %, PR = 1.66). Previously, it was found that the coverage/complementation of health expenses by SUS for older people was lower than for the youngest age group (16 to 29 years), although the demand for the service had similar proportions, denoting inequality in coverage that favors the youngest. As a consequence, the costs of households with an elderly person are higher, probably due to health plans or medicines¹⁹.

On this topic, it was also found that families with higher incomes and educational levels spent more money with health care, as well as white-skinned people and household heads in a stable union. According to Silveira et al.²², one of the explanations for the absence of health expenses in families with low socioeconomic levels may be the existence of SUS. However, “when health expenses exist, they pose a much greater impact on the incomes of the poorest families, which characterizes them as highly regressive”²², and it is the result of inequality in income distribution. Thus, the means of financing health costs is still a factor that promotes inequity²³, since the weight of income expenditures of families with better socioeconomic levels is very small, evidencing a huge redistributive potential and impact on poverty²².

Also important to point out that a large portion (46%) of people with indications of CMD who reported morbidity in the 15 days prior to survey did not seek care or health professionals. In Brazil, the prevalence of people who are cared for when they seek medical care is high (around 98%), but access barriers seem to precede the search²⁴. In this study, given the needs of people who reported morbidity and did not seek health services, one should consider that the high rate of non-demand may reflect, in addition to individual choice, factors related to the system’s characteristics and service provision, as well as sociodemographic conditions that function as barriers to access.

CONCLUSION

The concept of health equity acknowledges the particularities and vulnerabilities of individuals or groups and encompasses the idea that the role of a system is to minimize inequalities in sickness and death, providing equal conditions for all. Returning to the initial objective of this study, and considering both the limitations and results found, one can grasp an understanding of how the public system has contributed to the reduction of health inequalities arising from social inequalities.

In 2008, among the population with indications of CMD (who reported morbidity in the 15 days prior to survey), people who went to a health service for assistance were able to access and be provided with care, regardless of its public or private nature. This finding shows that when the health service is sought after, there is no inequality in access in favor of people with better socioeconomic levels, which reflects a a situation of equity.

Another situation of equity seen refers to people with lower per capita income and educational level seeking SUS the most and having wider coverage and lower expenses when compared to people with higher socioeconomic statuses, which shows a compensation for existing social inequalities in health. However, the means of health financing still promotes inequity, because the impact on the poorest segments’ income is higher compared to more favored social classes in spite of their lower expenses, according to the existing literature.

In this study, a considerable portion of people with indications of CMD who reported morbidity did not seek care at a health service. Due to methodological issues, it was not possible to specify the reasons of their not seeking help in the face of morbidities, which opens the way for further research to complement the results found. However, the literature points out that this is usually linked to iniquities preceding the search, making unnecessary and avoidable inequalities in health care clear.

Accessibility remains more favorable to individuals with better living conditions, who do not experience the difficulties that become access barriers to the poorest. In addition, people of higher social strata have a varied offer of services in the private system and the possibility of duplicate coverage in certain cases, as observed among the youngest in this study.

Another inequality pointed out relates to coverage according to age. The youngest had more SUS coverage and lower expenses, even though they searched the system in the same proportion as other age groups, which shows that SUS’ offer is better for this population. Accordingly, expenses with health amidst the older population are higher, a situation of inequity of extreme relevance when one considers the current demographic and epidemiological transition. It is worth remembering that the elderly population is at greater risk of having CMD, since they have to deal with growing limitations, economic and social role changes, among others.

In view of the above, health expenses coverage/complementation by the SUS in the city of São Paulo, in 2008, favored the most needed portion of population, contributing to greater equality in the health-disease process. However, without disregarding its importance, this is only a small piece of the whole puzzle that composes equity. One cannot speak of a truly equitable health system if countless factors still underwrite the maintenance of social inequalities that function as factors determining countless diseases. Important to put the spotlight on the problems: of health service distribution, which do not include the whole population, as exemplified here by the older age group in the sample and penalize the lower social strata; and of the public and private systems’ role in the current model, which privileges people with better socioeconomic levels, for example, allowing double entry in the network.

Further studies highlighting the gaps and inefficiencies of the health system are important because they can help monitor performance and plan interventions for its improvement.

REFERÊNCIAS

• ¹
  Brasil. Constituição da República Federativa do Brasil. Brasília: Senado Federal; 1988. 292 p.
• ²
  Souza R. Políticas e práticas de saúde e equidade. Rev Esc Enferm USP. 2007; 41(esp.): 765-70. http://dx.doi.org/10.1590/S0080-62342007000500004
  » http://dx.doi.org/10.1590/S0080-62342007000500004
• ³
  Whitehead M. The concepts and principles of equity and health. Copenhague: World Health Organization; 2000.
• ⁴
  Viana ALD, Fausto MCR, Lima LD. Política de saúde e equidade. São Paulo Perspec. 2003; 17(1): 58-68. http://dx.doi.org/10.1590/S0102-88392003000100007
  » http://dx.doi.org/10.1590/S0102-88392003000100007
• ⁵
  Buss PM, Pellegrini Filho A. A Saúde e seus Determinantes Sociais. Physis. 2007; 17(1): 77-93. http://dx.doi.org/10.1590/S0103-73312007000100006
  » http://dx.doi.org/10.1590/S0103-73312007000100006
• ⁶
  Travassos C. Equidade e o Sistema Único de Saúde: uma contribuição para debate. Cad Saúde Pública. 1997; 13(2): 325-30. http://dx.doi.org/10.1590/S0102-311X1997000200024
  » http://dx.doi.org/10.1590/S0102-311X1997000200024
• ⁷
  Campos GWS. Reflexões Temáticas sobre Equidade e Saúde: o caso do SUS. Saúde Sociedade. 2006; 15(2): 23-33. http://dx.doi.org/10.1590/S0104-12902006000200004
  » http://dx.doi.org/10.1590/S0104-12902006000200004
• ⁸
  Silva ATC, Menezes PR. Esgotamento profissional e transtornos mentais comuns em agentes comunitários de saúde. Rev Saúde Pública. 2008; 42(5):921-9.
• ⁹
  Goldberg D, Huxley P. Perturbações mentais comuns: um modelo bio-social. Lisboa: Climepsi Editores; 1996.
• ¹⁰
  Lima MCP. Transtornos mentais comuns e uso de álcool na população urbana de Botucatu - SP: um estudo de co-morbidade e utilização de serviços [tese]. São Paulo: Faculdade de Medicina, Universidade de São Paulo; 2004
• ¹¹
  Viana MC, Andrade LH. Lifetime Prevalence, age and gender distribution and age-of-onset of psychiatric disorders in the São Paulo Metropolitan Area, Brazil: results from the São Paulo Megacity Mental Health Survey. Rev Bras Psiquiatr [Internet]. 2012 [citado em 29 set. 2017]; 34(3): 249-60. Disponível em: http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1516-44462012000300004&lng=en http://dx.doi.org/10.1016/j.rbp.2012.03.001
  » https://doi.org/10.1016/j.rbp.2012.03.001» http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1516-44462012000300004&lng=en
• ¹²
  Moraes Júnior EC. Prevalência e fatores de risco para transtorno mental comum na população urbana da região metropolitana de São Paulo, Brasil [dissertação]. Botucatu: Universidade Estadual Paulista "Júlio de Mesquita Filho"; 2010.
• ¹³
  Pinheiro KAT. Prevalência e fatores associados a transtornos mentais comuns (TMC) em adolescentes entre 15 e 18 anos de zona urbana do sul do Brasil [dissertação]. Pelotas: Universidade Católica de Pelotas; 2003.
• ¹⁴
  Santos KOB, Araújo TM, Pinho PS, Silva ACC. Avaliação de um instrumento de mensuração de morbidade psíquica: estudo de validação do Self-Reporting Questionnaire (SRQ-20). Rev Baiana Saúde Pública. 2010; 34(3): 544-60.
• ¹⁵
  Mari JJ, Williams P. A validity study of a Psychiatric Screening Questionnaire (SRQ-20) in Primary care in the city of São Paulo. Br J Psychiatry. 1986; 148(1): 23-6.
• ¹⁶
  Scazufca M, Menezes PR, Vallada H, Araya R. Validity of the self reporting questionnaire-20 in epidemiological studies with older adults. Soc Psychiatry Psychiatr Epidemiol. 2009; 44(3): 247-54. https://doi.org/10.1007/s00127-008-0425-y
  » https://doi.org/10.1007/s00127-008-0425-y
• ¹⁷
  Silva ZP, Ribeiro MCSA, Barata RB, Almeida MF. Perfil sociodemográfico e padrão de utilização dos serviços de saúde do Sistema Único de Saúde (SUS), 2003-2008. Ciênc Saúde Coletiva. 2011; 16(9): 3807-16. http://dx.doi.org/10.1590/S1413-81232011001000016
  » http://dx.doi.org/10.1590/S1413-81232011001000016
• ¹⁸
  Bós AMG, Bós AJG. Determinantes na escolha entre atendimento de saúde privada e pública por idosos. Rev Saúde Pública. 2004; 38(1): 113-20. http://dx.doi.org/10.1590/S0034-89102004000100016
  » http://dx.doi.org/10.1590/S0034-89102004000100016
• ¹⁹
  Medici A. Propostas para melhorar a cobertura, a eficiência e a qualidade no setor saúde. In: Bacha EL, Schwartzman S (Eds.). Brasil: a nova agenda social. Rio de Janeiro: Instituto de Estudos de Política Econômica-Casa das Garças; 2011. p. 23-93.
• ²⁰
  Louvison MCP, Lebrão AL, Duarte YAO, Santos JLF, Malik, AM, Almeida ES. Desigualdades no uso e acesso aos serviços de saúde entre idosos do município de São Paulo. Rev Saúde Pública. 2008; 42(4): 733-40. http://dx.doi.org/10.1590/S0034-89102008000400021
  » http://dx.doi.org/10.1590/S0034-89102008000400021
• ²¹
  Granja GF, Zoboli ELCP, Fortes PAC, Fracolli LA. Equidade no sistema de saúde brasileiro: uma teoria fundamentada em dados. Rev Baiana Saúde Pública. 2010; 34(1): 72-86. https://doi.org/10.22278/2318-2660.2010.v34.n1.a14
  » https://doi.org/10.22278/2318-2660.2010.v34.n1.a14
• ²²
  Silveira FG, Osorio RG, Piola SF. Os gastos das famílias com saúde. In: Silveira FG, Servo LM, Menezes T, Piola SF (Eds.). Gasto e consumo das famílias brasileiras contemporâneas. Brasília: Ipea; 2006.
• ²³
  Ugá MAD, Santos IS. Uma análise da equidade do financiamento do sistema de saúde brasileiro [relatório de pesquisa]. Rio de Janeiro: Escola Nacional de Saúde Pública, ENSP/FIOCRUZ; 2005.
• ²⁴
  Pinheiro RS, Viacava F, Travassos C, Brito AS. Gênero, morbidade, acesso e utilização de serviços de saúde no Brasil. Ciênc Saúde Coletiva. 2002; 7(4): 687-707. http://dx.doi.org/10.1590/S1413-81232002000400007
  » http://dx.doi.org/10.1590/S1413-81232002000400007

• Financial support: Coordination for the Improvement of Higher Education Personnel (CAPES), Ministry of Education.

Publication Dates

• Publication in this collection
  2018

History

• Received
  20 Jan 2017
• Reviewed
  05 Dec 2017
• Accepted
  05 Feb 2018