Abstract:

The study's objectives were to describe hospital admissions for primary care sensitive conditions (PCSCs) among residents of Itaboraí, a municipality in Rio de Janeiro State, Brazil, with a high rate of such admissions, to identify associated factors, and to explore the impact from closing a hospital. The ultimate aim was to contribute to the debate on the indicator's indiscriminate use for inferences on the expansion and quality of primary care. The study was based on data from the Hospital Information System of the Brazilian Unified National Health System. Associations between admissions for PCSCs and demographic and hospital-related variables were analyzed with the χ² test. Logistic regression models verified the year-by-year behavior of admissions for PCSCs. The most frequent admissions were for heart failure, asthma, gastroenteritis, chronic obstructive pulmonary disease, and diabetes mellitus. The odds of hospital admissions were higher in males, brown individuals, at age extremes, and in private and charity hospitals. The trend in the admissions rates for PCSCs over the years was influenced by the closing of a hospital, suggesting caution in attributing changes in admissions rates to quality improvement in primary care.

Hospitalization; Primary Health Care; Quality of Health Care

Introduction

Hospital admissions for primary care sensitive conditions (PCSCs) are admissions for diseases that are sensitive to control and reduction by means of accessible and effective primary care, involving prevention and continuity of care. They have been considered indirect indicators of quality of primary healthcare services, pointing to the possibility of barriers to accessing services (geographic, cultural, financial, or organizational) and amenable to monitoring the effectiveness and performance of the primary care system ¹1. Departamento de Atenção Básica, Secretaria de Atenção à Saúde, Ministério da Saúde. III Seminário Internacional Atenção Primária Saúde da Família: expansão com qualidade & valorização de resultados. Relatório das atividades. Brasília: Ministério da Saúde; 2008. (Série D. Reuniões e Conferências).^,22. Alfradique MA, Bonolo PF, Dourado I, Lima-Costa MF, Macinko J, Mendonça CS, et al. Internações por condições sensíveis à atenção primária: a construção da lista brasileira como ferramenta para medir o desempenho do sistema de saúde (Projeto ICSAP - Brasil). Cad Saúde Pública 2009; 25:1337-49.. Following the initiative of many countries in the previous decade, Brazil published its first List of Primary Care Sensitive Conditions in 2008, after independent initiatives in the states of Ceará (2001), Minas Gerais (2007), and Paraná (2006), not mutually comparable or appropriate for national use due to their regional specificities ²2. Alfradique MA, Bonolo PF, Dourado I, Lima-Costa MF, Macinko J, Mendonça CS, et al. Internações por condições sensíveis à atenção primária: a construção da lista brasileira como ferramenta para medir o desempenho do sistema de saúde (Projeto ICSAP - Brasil). Cad Saúde Pública 2009; 25:1337-49.. The purpose of the Brazilian list was to create an instrument that could be used for evaluating primary care in the country, contributing to health services planning and management at the national, state, and municipal levels.

In order to better understand admissions for PCSCs, Brazilian and international studies have frequently focused on their characterization and the identification of associated socio-demographic and healthcare-related variables ³3. Magan P, Otero A, Alberquilla A, Ribera JM. Geographic variations in avoidable hospitalizations in the elderly, in a health system with universal coverage. BMC Health Serv Res 2008; 8:42.^,44. Carter MW, Datti B, Winters JM. ED visits by older adults for ambulatory care-sensitive and supply-sensitive conditions. Am J Emerg Med 2006; 24:428-34.^,55. Laditka JN, Laditka SB, Mastanduno MP. Hospital utilization for ambulatory care sensitive conditions: health outcome disparities associated with race and ethnicity. Soc Sci Med 2003; 57:1429-41.^,66. Nayar P, Nguyen AT, Apenteng B, Yu F. Preventable hospitalizations: does rurality or non-physician clinician supply matter? J Community Health 2012; 37:487-94.^,77. Billings J, Zeitel L, Lukomnic J, Carey TS, Blank AE, Newman L. Impact of socioeconomic status on hospital use in New York City. Health Aff (Millwood) 1993; 12:162-73.^,88. Blustein J, Hanson K, Shea S. Preventable hospitalizations and socioeconomic status. Health Aff (Millwood) 1998; 17:177-89.^,99. Nedel FB, Facchini LA, Martín-Mateo M, Vieira LAS, Thumé E. Programa Saúde da Família e condições sensíveis à atenção primária, Bagé - RS. Rev Saúde Pública 2008; 42:1041-52.^,1010. Junqueira RMP, Duarte EC. Internações hospitalares por causas sensíveis à atenção primária no Distrito Federal, 2008. Rev Saúde Pública 2012; 46:761-8.^,1111. Dias-da-Costa JS, Borba LG, Pinho MN, Chatkin M. Qualidade da atenção básica mediante internações evitáveis no Sul do Brasil. Cad Saúde Pública 2008; 24:1699-707.^,1212. Moura BLA, Cunha RC, Aquino R, Medina MG, Mota ELA, Macinko J, et al. Principais causas de internação por condições sensíveis à atenção primária no Brasil: uma análise por faixa etária e região. Rev Bras Saúde Matern Infant 2010; 10 Suppl 1:583-91.^,1313. Caldeira AP, Fernandes VBL, Fonseca WP, Faria AA. Internações pediátricas por condições sensíveis à atenção primária em Montes Claros, Minas Gerais, Brasil. Rev Bras Saúde Matern Infant 2011; 11:61-71.^,1414. Oliveira BRG, Vieira CS, Collet N, Lima RAG. Causas de hospitalização no SUS de crianças de zero a quatro anos no Brasil. Rev Bras Epidemiol 2010; 13:268-77.^,1515. Boing AF, Vicenzi RB, Magajewski F, Boing AC, Moretti-Pires RO, Peres KG, et al. Redução das internações por condições sensíveis à atenção primária no Brasil entre 1998-2009. Rev Saúde Pública 2012; 46:359-66.^,1616. Pazó RG, Frauches DO, Galvêas DP, Stefenoni AV, Cavalcante ELB, Pereira-Silva FH. Internações por condições sensíveis à atenção primária no Espírito Santo: estudo ecológico descritivo no período 2005-2009. Epidemiol Serv Saúde 2012; 21:275-82.^,1717. Costa JSD, Reis MC, Silveira Filho CV, Linhares RS, Piccinini F, Fantinel E. Prevalência de internação hospitalar e fatores associados em Pelotas - RS. Rev Saúde Pública 2010; 44:923-33.^,1818. Fernandes VBL, Caldeira AP, Faria AA, Rodrigues Neto JF. Internações sensíveis na atenção primária como indicador de avaliação da Estratégia Saúde da Família. Rev Saúde Pública 2009; 43:928-36.. Brazil has also seen analytical ecological studies that have associated the decline in admissions for PCSCs with the expansion of primary care by the Family Health Strategy (FHS) ¹⁹19. Macinko J, Dourado I, Aquino R, Bonolo PF, Lima-Costa MF, Medina MG, Mota E, et al. Major expansion of primary care in Brazil linked to decline in unnecessary hospitalization. Health Aff (Millwood) 2010; 29:2149-60.^,2020. Dourado I, Oliveira VB, Aquino R, Bonolo P, Lima-Costa MF, Medina MG, et al. Trends in primary health care-sensitive conditions in Brazil: the role of the Family Health Program (Project ICSAP-Brazil). Med Care 2011; 49:577-84., as well as descriptive studies on the correlation between the decline in the rates or proportions of admissions for PCSCs and the strategy's expansion in more limited geographic areas ²¹21. Barreto JOM, Nery IS, Costa MSC. Estratégia Saúde da Família e internações hospitalares em menores de 5 anos no Piauí, Brasil. Cad Saúde Pública 2012; 28:515-26.^,2222. Veloso RC, de Araújo MRN. Avaliação da resolutividade do Programa Saúde da Família em municípios de pequeno porte no Estado de Minas Gerais. Rev APS 2009; 12:238-43.^,2323. Rehem TCMSB, Egry EY. Internações por condições sensíveis à atenção primária no Estado de São Paulo. Ciênc Saúde Coletiva 2011; 16:4755-66..

The current study originated from a broader project involving epidemiological monitoring in the area reserved for implementation of the Rio de Janeiro State Petrochemical Complex (COMPERJ) ²⁴24. Toledo LM, Sabroza PC. Plano de Monitoramento Epidemiológico do Processo de Implantação do COMPERJ - Síntese anual 2010. Monitoramento da evolução de causas relevantes de adoecimento e morte nos municípios de Itaboraí, Guapimirim e Cachoeiras de Macacu: a situação epidemiológica em 2010. Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2011., the largest single project in the history of Petrobrás, launched in 2006 and scheduled to begin operations in 2014. The study was motivated by the observation of high admissions rates for PCSCs in Itaboraí, the municipality (county) where COMPERJ is located in Greater Metropolitan Area II, Rio de Janeiro. Estimates for the municipality in the years 2008 ²⁵25. Secretaria de Estado de Saúde e Defesa Civil. Saúde na área: regionalização. Caderno de Informações em Saúde do Estado do Rio de Janeiro. Rio de Janeiro: Secretaria de Estado de Saúde e Defesa Civil; 2009. and 2010 and 2011 (Índice de Desempenho do Sistema Único de Saúde. http://idsus.saude.gov.br/mapas.html, accessed on 23/Mar/2016) indicated admissions rates for PCSCs of 41.6%, 43.1%, and 48.2%, respectively, far higher than the overall rates in the region of Rio de Janeiro State where the municipality is located, i.e., slightly over 30%.

Itaboraí has a territory of 429km² and a population of 218,008 as of 2010 (Instituto Brasileiro de Geografia e Estatística. http://www.sidra.ibge.gov.br/pnad/default.asp, accessed on 23/Mar/2016). The population is almost entirely urban (98.8% in 2000), with a mean population density of just over 500 inhabitants/km², while exceeding 1,000 inhabitants/km² in more central urban areas ²⁶26. Secretaria Municipal de Planejamento e Coordenação, Prefeitura Municipal de Itaboraí. Cadernos ITADADOS 2005/2006. Itaboraí: Prefeitura Municipal de Itaboraí; 2007.. According to data from the Brazilian National Household Sample Survey (PNAD), in 2010 only 30% of households were connected to the public water supply system, 40.8% were connected to the public sewage system or to storm drains, and 95.9% received regular garbage collection (Instituto Brasileiro de Geografia e Estatística. http://www.sidra.ibge.gov.br/pnad/default.asp, accessed on 23/Mar/2016). The municipality has already experienced an epidemiological transition, and the four leading causes of death are diseases of the circulatory system (23%), external causes (13%), diseases of the respiratory system - especially influenza, pneumonia, and chronic obstructive pulmonary disease (COPD) - (11%), and endocrine diseases (11%) ²⁴24. Toledo LM, Sabroza PC. Plano de Monitoramento Epidemiológico do Processo de Implantação do COMPERJ - Síntese anual 2010. Monitoramento da evolução de causas relevantes de adoecimento e morte nos municípios de Itaboraí, Guapimirim e Cachoeiras de Macacu: a situação epidemiológica em 2010. Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2011..

Primary health care in the municipality was initially organized in the 1990 and was consolidated with the Family Health Program (FHP)/FHS in the 2000s, with some 45 teams - each consisting of one physician, nurse, and nurse technician and six community health workers - distributed across 34 units, with approximately 70% population coverage (Ministério da Saúde. http://dab.saude.gov.br/portaldab/historico_cobertura_sf.php, accessed on 23/Mar/2016). From 2009 to 2012 there were problems with administrative discontinuity, insufficient resources, and low capacity for coordination of the health teams' work ²⁷27. Mudjalieb AA. Produção de informação e conhecimento e práticas de participação popular na Estratégia Saúde da Família: o caso do município de Itaboraí, RJ [Dissertação de Mestrado]. Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2011.. In 2012, the performance of the Brazilian Unified National Health System (SUS) in the municipality was assessed as precarious (4.95, on a scale from 0 to 10), close to that of the Rio de Janeiro State as a whole (4.58) and below the national average (5.47) (Índice de Desempenho do Sistema Único de Saúde. http://idsus.saude.gov.br/mapas.html, accessed on 23/Mar/2016).

This article aims to analyze hospital admissions for PCSCs among residents of Itaboraí, identify factors associated with their occurrence among all admissions of residents in the municipality covered by the SUS, and explore the impact of closing a hospital on admissions rates for PCSCs, thereby contributing to an on-going debate on the indiscriminate use of this indicator for inferences on the expansion and quality of primary care ²⁸28. Rodrigues-Bastos RM, Campos EMS, Ribeiro LC, Bastos Filho MG, Bustamante-Teixeira MT. Internações por condições sensíveis atenção primária, Minas Gerais, 2000 e 2010. Rev Saúde Pública 2014; 48:958-67.^,2929. Lugo-Palacios DG, Cairns J. Using ambulatory care sensitive hospitalisations to analyse the effectiveness of primary care services in Mexico. Soc Sci Med 2015; 144:59-68.^,3030. Gonçalves MR, Hauser L, Prestes IV, Schmidt MI, Duncan BB, Harzheim E. Primary health care quality and hospitalizations for ambulatory care sensitive conditions in the public health system in Porto Alegre, Brazil. Fam Pract 2016; 33:238-42.. The target period is 2006 to 2011, when coverage by the FHP/FHS in the municipality varied irregularly from approximately 65% to 71% (Índice de Desempenho do Sistema Único de Saúde. http://idsus.saude.gov.br/mapas.html, accessed on 23/Mar/2016).

Methods

The study was based on secondary data from the Hospital Information System of the SUS (SIH/SUS), including all hospital admissions covered by the SUS for residents of the municipality of Itaboraí, Rio de Janeiro State, from 2006 to 2011, excluding cases involving long-term hospitalization or childbirth.

A databank was built by linkage of files from reduced Authorizations for Hospital Admissions (AIH), available on the DATASUS website (http://www.datasus.gov.br). The electronic files, obtained in zipped dbc format, were combined and unzipped in dbf format, using TabWin from DATASUS.

In building the databank, we selected variables on the hospitals (identification, type, and municipality), patients (date of birth, sex, and race/color), and admissions (admission date, discharge date, principal diagnosis, outcome, amount paid by the SUS, and amount paid for use of the intensive care unit - ICU). The analyses here refer to two three-year periods, 2006-2008 and 2009-2011.

For the overall characterization of admissions for PCSCs, the respective hospitalization rates were calculated as the number of patients admitted for diagnoses classified by the Brazilian Ministry of Health Directive 221/MS/SAS of April 17, 2008 ³¹31. Secretaria de Atenção à Saúde, Ministério da Saúde. Portaria nº 221, de 17 de abril de 2008. Diário Oficial da União 2008; 18 abr., as PCSCs in the municipality divided by the municipality's total population, multiplied by 10,000; the municipality's population was obtained from the Brazilian Institute of Geography and Statistics (IBGE).

Bivariate analyses were performed to explore associations between admissions for PCSCs (versus other causes) and potential explanatory variables for the variation, available in the databank. Since the independent variables were categorized, we used the chi-square test (χ²), verifying the null hypothesis of non-association. Logistic regression models were explored to identify the explanatory variables' independent effect on admissions for PCSCs, opting to consider possible differences in their odds year by year, with 2009 as the reference. All statistical analyses used SAS version 9.2 (SAS Inst., Cary, USA).

The research project was approved by the Institutional Review Board of National School of Public Health, Oswaldo Cruz Foundation (case review number 09410712.7.0000.5240).

Results

Excluding long-term hospitalizations and admissions for childbirth, there were 38,702 and 30,910 admissions covered by the SUS for residents of Itaboraí in 2006-2008 and 2009-2011, of which 39.2% and 32.4% were for PCSCs, respectively. In both three-year periods, patients admitted for PCSCs were significantly older on average (2006-2008, 48.3 ± 27, 3 years; 2009-2011, 44.0 ± 29, 1 year) than those admitted for other causes (2006-2008, 34.5 ± 22, 6 years; 2009-2011, 35.0 ± 22, 7 years). Admissions for PCSCs also showed lower mean cost and narrower range (2006-2008, BRL 447.90 ± 376.70; 2009-2011, BRL 546.60 ± 525.10) compared to admissions for other causes (2006-2008, BRL 574.50 ± 1,154.30; 2009-2011, BRL 742.50 ± 1,902.80).

Figure 1 shows the trend in rates per 10,000 inhabitants for admissions due to PCSCs, versus admissions for other causes, over the six years of study, indicating a downward trend in both from 2006 to 2009, with 2010 showing a peak increase in admissions for other causes and an abrupt drop in admissions for PCSCs. During the study period, the admissions rate for PCSCs dropped gradually from 234.7 to 179.8/10,000 inhabitants, from 2006 to 2010, while from 2010 to 2011 it showed a sharp drop, reaching 60 admissions per 10,000 inhabitants. The admissions rate for other causes showed a more gradual drop, ranging between approximately 360 and 300 admissions/10,000 inhabitants.

As shown in Table 1, Itaboraí showed a highly concentrated distribution pattern in admissions for PCSCs according to causes, with the first five accounting for 69.7 and 62.7% of the total in 2006-2008 and 2009-2011, respectively. Heart failure was the leading cause of admissions for PCSCs in both three-year periods, followed by asthma, gastroenteritis, other COPD, and diabetes mellitus. Despite the drop in the absolute number of admissions for the five causes, the fact stands out that they are the same ones, and in the same order, in the two three-year periods. From the first three-year period to the second, there was a drop in the relative frequency of heart failure and an increase in that of asthma. However, there was an absolute drop in both.

Among admissions for PCSCs in persons 60 years or older, there was a striking concentration of chronic diseases, especially heart failure, stroke, COPD, hypertension, and diabetes mellitus. Meanwhile, in children 1 to 4 years of age there was a concentration of admissions for gastroenteritis, bacterial pneumonia, and asthma.

Table 2 explores possible associations between admissions for PCSCs (versus admissions for other causes) and socio-demographic and hospital-related variables. Both periods showed more admissions for PCSCs in persons 1 to 4 years of age or 60 and older, as well as in males. The race variable was only included recently on SIH/SUS, and the data from the first three-year period are practically invalidated by the high proportion of cases missing this information. However, from 2009 to 2011 the admissions rates for PCSCs were higher in individuals classified as brown (mixed-race) (40.6%). Table 2 further shows the concentration of admissions for PCSCs in private hospitals outsourced by the SUS and in the municipality of Itaboraí itself. There was a clear drop in these admissions from the first period to the second in outsourced hospitals, but maintaining their proportion in total admissions at greater than 65%.

Table 3 shows two models that identify the independent effect of factors associated with admissions for PCSCs in Itaboraí from 2006 to 2011, considering inclusion versus non-inclusion of the independent variable "administrative classification" (i.e., of the hospital, whether outsourced, charitable, etc.). Both models excluded individuals classified as Asian-descendant or indigenous (due to their small numbers). The hospital's administrative classification had a major impact on the estimates, comparing outsourced, charitable, and other hospitals, particularly showing a change in the behavior of the indicators for year of hospitalization. Model 1, which did not adjust for the hospital's administrative classification, maintained a consistently downward gradient in the odds of admissions for PCSCs, while in model 2, which did not make this adjustment, the gradient was not maintained.

The results in model 2 show that considering the age variable and taking 18 to 59 years as the reference age bracket, the odds of occurrence of admissions for PCSCs (versus non-occurrence) were higher among younger individuals (< 1 year, OR = 1.493; 1-4 years, OR = 1.727; 5-17 years, OR = 1.122) and especially among the elderly (60-69 years, OR = 3.336; 70-79 years, OR = 3.570; ≥ 80 years, OR = 3.819). Maintaining the other variables constant, the odds of admissions for PCSCs were 11.9% lower in females and 8.7% higher in those classified as brown, when compared, respectively, to males and whites. Considering the hospital's administrative classification, admissions for PCSCs were higher in private outsourced and charitable hospitals. Finally, the multivariate analysis showed admissions rates for PCSCs among Itaboraí residents hospitalized in the municipality itself, and higher odds of admissions for PCSCs in 2009, taken as the reference year, and in which the admissions rate for PCSCs only failed to differ statistically from the rate for 2011.

Discussion

An important contribution by this study and that corroborates other recent studies is the need for caution when interpreting results and making inferences associating a reduction in admissions for PCSCs with success in primary care interventions such as the FHS, although more robust ecological studies may be relevant and efficient in more global evaluations ¹⁹19. Macinko J, Dourado I, Aquino R, Bonolo PF, Lima-Costa MF, Medina MG, Mota E, et al. Major expansion of primary care in Brazil linked to decline in unnecessary hospitalization. Health Aff (Millwood) 2010; 29:2149-60.^,2020. Dourado I, Oliveira VB, Aquino R, Bonolo P, Lima-Costa MF, Medina MG, et al. Trends in primary health care-sensitive conditions in Brazil: the role of the Family Health Program (Project ICSAP-Brazil). Med Care 2011; 49:577-84.. In more localized studies, it is essential to grasp contextual issues, sometimes highly specific, that can help explain the event. A superficial observation of the results obtained here could easily suggest an association between the drop in admissions for PCSCs and expansion of the FHP/FHS in Itaboraí, starting in the 1990s and covering 71.2% of the population in 2011 according to data from the Ministry of Health's Department of Primary Care (http://dab.saude.gov.br/portaldab/historico_cobertura_sf.php, accessed on 23/Mar/2016). Other factors clearly intervened, although it was not possible to analyze socioeconomic conditions or distribution and access to services by the population, also potentially relevant. Although the paradigm shift in the organization of the healthcare model through the implementation and expansion of the FHP/FHS since the 1990s reorganized and upgraded health services in Itaboraí, it would not be possible to correlate the drop in admissions for PCSCs with the improvement in healthcare.

Various authors have identified limitations in the use of admissions for PCSCs as an indicator of quality of primary care, depending on the culture of use of health services by the study population, hospital admissions policy, morbidity and disease prevalence, individual factors related to hospitalization, and the very definition of "primary care sensitive" diseases ¹1. Departamento de Atenção Básica, Secretaria de Atenção à Saúde, Ministério da Saúde. III Seminário Internacional Atenção Primária Saúde da Família: expansão com qualidade & valorização de resultados. Relatório das atividades. Brasília: Ministério da Saúde; 2008. (Série D. Reuniões e Conferências).^,3232. Caminal Homar J, Casanova Matutano C. La evaluación de la atención primaria y las hospitalizaciones por ambulatory care sensitive conditions. Marco conceptual. Aten Primaria 2003; 31:61-5.. The indicator's use for inferences on the expansion and quality of care does not eliminate the need to consider factors related to health services organization and practice. Rodrigues-Bastos et al. ²⁸28. Rodrigues-Bastos RM, Campos EMS, Ribeiro LC, Bastos Filho MG, Bustamante-Teixeira MT. Internações por condições sensíveis atenção primária, Minas Gerais, 2000 e 2010. Rev Saúde Pública 2014; 48:958-67., in an ecological study, found no correlation between areas with expansion of coverage by the FHP/FHS and a decrease in admissions for PCSCs in municipalities in Minas Gerais State, Brazil, from 2000 to 2010. A study in Mexico ²⁹29. Lugo-Palacios DG, Cairns J. Using ambulatory care sensitive hospitalisations to analyse the effectiveness of primary care services in Mexico. Soc Sci Med 2015; 144:59-68. concluded that admissions for PCSCs were not a valid measure of the effectiveness of primary care in the country from 2001 to 2011, given the expansion of health insurance coverage. However, the authors confirmed the indicator's applicability for comparing different regions or to make regional comparisons, as long as health insurance coverage was considered and allowed the identification of possible differences in provision of primary health care services. Gonçalves et al. ³⁰30. Gonçalves MR, Hauser L, Prestes IV, Schmidt MI, Duncan BB, Harzheim E. Primary health care quality and hospitalizations for ambulatory care sensitive conditions in the public health system in Porto Alegre, Brazil. Fam Pract 2016; 33:238-42., using a validated instrument for evaluation of primary health care - PCATool Brazil - in a cohort study in Porto Alegre, Rio Grande do Sul State, found that improvement in the supply of public healthcare services did not influence the admissions rate for PCSCs in a context of previously limited services supply. Finally, a study in Espírito Santo State, Brazil, using bivariate analysis, found a positive association between coverage of the FHP/FHS and admissions for PCSCs. According to the authors, this result could be explained by the prior scarcity of access to (and availability of) healthcare services; to the extent that services began to be supplied, a brief increase in hospitalizations would result from serving a repressed demand ³³33. Pazó RG, Frauches DO, Molina MCB, Cade NV. Modelagem hierárquica de determinantes associados a internações por condições sensíveis à atenção primária no Espírito Santo, Brasil. Cad Saúde Pública 2014; 30:1891-902..

In light of all the above, this study's results indicated a downward trend in hospital admissions in Itaboraí, as well as a decline in admissions for PCSCs, which accounted for 39.2% of all admissions in 2006-2008 and 32.4% in 2009-2011.

When transformed into rates, both admissions for PCSCs and admissions for other causes in Itaboraí declined over the course of the six years. However, admissions rates for PCSCs experienced a sharp drop in 2010-2011, which appears to be explained largely by the closing of a hospital outsourced by the SUS and which had received a large share of these admissions in the municipality ³⁴34. Toledo LM, Sabroza PC. Plano de Monitoramento Epidemiológico do Processo de Implantação do COMPERJ - síntese anual 2011. Monitoramento da evolução dos indicadores de saúde e segurança nos municípios de Itaboraí, Guapimirim e Cachoeiras de Macacu e no Distrito de Monjolos (São Gonçalo). Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2012..

The downward trend in total hospital admissions and in admissions rates for PCSCs has been shown in many studies published in Brazil.

Another study also based on data from SIH/SUS, from 1998 to 2009, found a 3.7% annual reduction in the admissions rates for PCSCs in Brazil ¹⁵15. Boing AF, Vicenzi RB, Magajewski F, Boing AC, Moretti-Pires RO, Peres KG, et al. Redução das internações por condições sensíveis à atenção primária no Brasil entre 1998-2009. Rev Saúde Pública 2012; 46:359-66.. Other studies in Brazil found a 15.8% drop in hospitalizations for PCSCs and 10.1% for other causes, respectively, from 2000 to 2006, and a 5% decline in admissions rates for PCSCs from 1999 to 2007 ²2. Alfradique MA, Bonolo PF, Dourado I, Lima-Costa MF, Macinko J, Mendonça CS, et al. Internações por condições sensíveis à atenção primária: a construção da lista brasileira como ferramenta para medir o desempenho do sistema de saúde (Projeto ICSAP - Brasil). Cad Saúde Pública 2009; 25:1337-49.^,3535. Macinko J, Oliveira VB, Turci MA, Guanais FC, Bonolo P, Lima-Costa MF. The influence of hospital and primary care supply on ambulatory care sensitive hospitalizations among adults in Brazil, 1999-2007. Am J Public Health 2011; 101:1963-70..

The more abrupt drop in admissions for PCSCs observed in this study, largely related to the closing of a hospital, raises doubts concerning both the real need for a large share of the previous admissions (which appears not to have been redistributed) and the data's trustworthiness. The multivariate analysis, by considering the effect of private hospitals outsourced by the SUS, suggests an increase in the odds of admissions for PCSCs as part of overall admissions from 2006 to 2009, a drop in 2010, and an admissions rate for PCSCs for 2011 that was not statistically different from the rate in 2009.

The three leading groups according to the International Classification of Diseases, 10th revision (ICD-10), observed in Itaboraí in the two three-year periods were heart failure, asthma, and gastroenteritis, with a reduction over time in the first and third groups and an increase in the relative frequency of asthma from 2006 to 2011. Importantly, there was a pattern of high rates in certain diagnostic groups that were well-established in the municipality over the course of both study periods.

Other studies have addressed the importance of heart failure among overall admissions for PCSCs in Brazil, although the rates observed here were far higher. Heart failure was identified as the second leading cause of admissions for PCSCs in the country, with 16.8 per 10,000 inhabitants in 2006 ²2. Alfradique MA, Bonolo PF, Dourado I, Lima-Costa MF, Macinko J, Mendonça CS, et al. Internações por condições sensíveis à atenção primária: a construção da lista brasileira como ferramenta para medir o desempenho do sistema de saúde (Projeto ICSAP - Brasil). Cad Saúde Pública 2009; 25:1337-49.. Heart failure was also identified as the second cause of admissions for PCSCs in a study in the Federal District (Brasilia), with 14.4 admissions per 10,000 inhabitants in 2008, mainly in males and in the 60 and older age bracket ¹⁰10. Junqueira RMP, Duarte EC. Internações hospitalares por causas sensíveis à atenção primária no Distrito Federal, 2008. Rev Saúde Pública 2012; 46:761-8.. A study from 1995 to 2004 in Pelotas, Rio Grande do Sul State, found that heart failure was the leading cause of admissions for PCSCs in males during eight of the ten years studied ¹⁷17. Costa JSD, Reis MC, Silveira Filho CV, Linhares RS, Piccinini F, Fantinel E. Prevalência de internação hospitalar e fatores associados em Pelotas - RS. Rev Saúde Pública 2010; 44:923-33..

Asthma appeared as the second leading group of admissions for PCSCs in Itaboraí. Importantly, the disease has become a major public health problem, with increasing incidence in the last three decades and 15-20% prevalence rates in Australia, England, New Zealand, Ireland, Brazil, and Costa Rica ³⁶36. Kuschnir FC, Cunha AJLA, Braga DAC, Silveira HHN, Barroso MH, Aires ST. Asma em escolares de 13 e 14 anos do Município de Nova Iguaçu, Rio de Janeiro, Brasil: estimativas de prevalência, gravidade e diferenças de gênero. Cad Saúde Pública 2007; 23:919-26.. Many factors associated with asthma prevalence have been described in the literature: environmental allergens, increasing air pollution, climate variations, and smoking ¹²12. Moura BLA, Cunha RC, Aquino R, Medina MG, Mota ELA, Macinko J, et al. Principais causas de internação por condições sensíveis à atenção primária no Brasil: uma análise por faixa etária e região. Rev Bras Saúde Matern Infant 2010; 10 Suppl 1:583-91.. Admissions for asthma have been associated with female gender, age 60-69 years, low schooling, low family income, family history of asthma, personal atopy, smoking, and minor psychiatric disorders ³⁷37. Macedo SEC, Menezes AMB, Knorst M, Dias-da-Costa JS, Gigante DP, Olinto MTA, et al. Fatores de risco para a asma em adultos, Pelotas, Rio Grande do Sul, Brasil. Cad Saúde Pública 2007; 23:863-74.. Focusing on the pediatric population, factors associated with the development of this disease and described in the literature are genetic load, diet, stress, antibiotics, cesarean delivery, breastfeeding, exposure to animals, and family size and structure ³⁸38. Subbarao P, Mandhane PJ, Sears MR. Asthma: epidemiology, etiology and risk factors. CMAJ 2009; 181:E181-90.. An important risk factor for the high asthma rate in Itaboraí is pollution, since the municipality is historically famous for its brick and tile factories, and the new petrochemical complex tends to further increase such pollution.

Gastroenteritis was the third leading cause of admissions for PCSCs in Itaboraí during both periods. Studies in Espírito Santo (2005-2009) and Minas Gerais (1998-2004) that included gastroenteritis among the leading causes of admissions for PCSCs found that the disease was highly prevalent in areas with heavily concentrated poverty and precarious sanitation ¹⁶16. Pazó RG, Frauches DO, Galvêas DP, Stefenoni AV, Cavalcante ELB, Pereira-Silva FH. Internações por condições sensíveis à atenção primária no Espírito Santo: estudo ecológico descritivo no período 2005-2009. Epidemiol Serv Saúde 2012; 21:275-82.^,3939. Perpetuo IHO, Wong LR. Atenção hospitalar por condições sensíveis à atenção ambulatorial (CSAA) e as mudanças do seu padrão etário: uma análise exploratória dos dados de Minas Gerais. In: Anais do XII Seminário sobre a Economia Mineira. Belo Horizonte: Editora UFMG; 2006.. Precarious sanitation conditions are typical in Itaboraí, where an association has been reported between admissions for diarrhea in children and factors like disordered urban growth and increasing population density in peripheral urban and rural areas ⁴⁰40. Torres RMC, Bittencourt SA, Oliveira RM, Siqueira ASP, Sabroza PC, Toledo LM. Uso de indicadores de nível local para análise espacial da morbidade por diarreia e sua relação com as condições de vida. Ciênc Saúde Coletiva 2013; 18:1441-50..

Chronic non-communicable diseases weighed heavily as causes of admissions for PCSCs, showing the consistency between the current study's results and another study on these admissions in elderly residents of Rio de Janeiro State from 2000 to 2010, which particularly evidenced the relevance of chronic obstructive pulmonary disease and heart failure ⁴¹41. Marques AP. Análise das causas de internação de idosos segundo a classificação de condições sensíveis à atenção primária: estudo de evolução temporal no Estado do Rio de Janeiro, 2012 [Dissertação de Mestrado]. Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2012.. Chronic diseases are now high on the list of challenges for the organization and coordination of health services, assuming the essential role of effective primary care and including the detection of exposed populations, adequate treatment, and continuity of care in reducing hospitalizations.

According to the results, increased likelihood of admission for a PCSC in Itaboraí in the SUS was associated with: male gender, age extremes (especially 1-4 years and 60 years or older), self-reported brown skin color, hospitalization in outsourced or charitable hospitals in the SUS, and hospitalization in Itaboraí itself. The study did not prioritize the application of the theoretical model proposed by Nedel et al. ⁴²42. Nedel FB, Facchini LA, Bastos JL, Martín-Mateo M. Conceptual and methodological aspects in the study of hospitalizations for ambulatory care sensitive conditions. Ciênc Saúde Coletiva 2011; 16 Suppl 1:1145-54. to analyze the impact of primary care on admissions for PCSCs, which would have placed greater emphasis on the inclusion of health system performance indicators (proximal factors related to organization, structure, governability, and model of care). However, the results are relevant to the model, identifying the strong weight of the "administrative classification" variable, which certainly reflects relevant organizational differences. Among the socio-demographic variables, age was found to be extremely important, but gender and race appeared less so. The identification of more admissions due to PCSCs in males has been widely reported in the literature. A study in 2002 concluded that women use preventive health services more, while men are more like to use emergency services ⁴³43. Pinheiro RS, Viacava F, Travassos C, Brito AS. Gênero, morbidade, acesso e utilização de serviços de saúde no Brasil. Ciênc Saúde Coletiva 2002; 7:687-707.. Higher alcohol and drug use and smoking are other factors that expose men to greater lifetime health risk when compared to women ⁴⁴44. Jinez MLJ, Molina de Souza R, Pillon SC. Uso de drogas e fatores de risco entre estudantes de ensino médio. Rev Latinoam Enferm 2009; 17:246-52.. Marques ⁴¹41. Marques AP. Análise das causas de internação de idosos segundo a classificação de condições sensíveis à atenção primária: estudo de evolução temporal no Estado do Rio de Janeiro, 2012 [Dissertação de Mestrado]. Rio de Janeiro: Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz; 2012. found that the elderly male population in Rio de Janeiro State had a higher risk of hospitalization for PCSCs, especially due to COPD.

Generally speaking, the positive association between admissions for PCSCs and age extremes in this study has also been identified in other studies ⁹9. Nedel FB, Facchini LA, Martín-Mateo M, Vieira LAS, Thumé E. Programa Saúde da Família e condições sensíveis à atenção primária, Bagé - RS. Rev Saúde Pública 2008; 42:1041-52.^,1010. Junqueira RMP, Duarte EC. Internações hospitalares por causas sensíveis à atenção primária no Distrito Federal, 2008. Rev Saúde Pública 2012; 46:761-8.. Among younger individuals, this highlights the higher admissions rates for PCSCs in the 1 to 4 year age bracket, compared to infants (< 1 year). In the Federal District of Brazil, high admissions rates for PCSCs were found in infants (< 1 year), an important reduction in the subsequent age brackets (1 to 29 years), and then a gradual increase with age ³⁵35. Macinko J, Oliveira VB, Turci MA, Guanais FC, Bonolo P, Lima-Costa MF. The influence of hospital and primary care supply on ambulatory care sensitive hospitalizations among adults in Brazil, 1999-2007. Am J Public Health 2011; 101:1963-70.. A study in Bagé, Rio Grande do Sul State, also found higher admissions rates for PCSCs in age extremes, in both males and females ⁴⁴44. Jinez MLJ, Molina de Souza R, Pillon SC. Uso de drogas e fatores de risco entre estudantes de ensino médio. Rev Latinoam Enferm 2009; 17:246-52.. More specifically, another study found a statistically significant association between admissions for PCSCs and age 60 years or older ¹⁸18. Fernandes VBL, Caldeira AP, Faria AA, Rodrigues Neto JF. Internações sensíveis na atenção primária como indicador de avaliação da Estratégia Saúde da Família. Rev Saúde Pública 2009; 43:928-36..

Interpretations of the results pertaining to the "race" variable are somewhat complicated by the high rate of missing data in the first three-year period. However, the decision not to ignore the high rate of admissions for PCSCs among individuals with brown skin color is justified by the results' consistency, despite the limitations. Brown race is by far the most common, and together with blacks, brown are associated with less favorable socioeconomic conditions. The higher admissions rates for PCSCs among brown individuals, observed consistently here, can be explained by less favorable socioeconomic conditions and the resulting difficulties in access to health services. However, from this perspective the lower admissions rate for PCSCs among blacks when compared to whites is surprising, but did not reach statistical significance. This difference should be better explored and explained in future studies.

Finally, this study indicates that when other variables are maintained constant, admissions for PCSCs are concentrated in the age extremes (children and elderly), with a pattern featuring mostly chronic diseases in the elderly, but also gastroenteritis and bacterial pneumonias in children.

The study confirms the importance of exercising caution in inferences on the meaning of decreasing admissions rates for primary care-sensitive conditions in local situations, more limited and heavily context-dependent. The capacity for trustworthy inferences in ecological studies can be favored by a robust data set, dealing with a diversity of contexts and allowing compensations between realities and appreciation of average behavior. Meanwhile, at a more specific level, the study confirms the high rate of PCSCs in Itaboraí, with a pattern featuring chronic diseases in the elderly but also with a strong presence of gastroenteritis and bacterial pneumonias in children. Such results appear not only to reveal persistent problems in the health system, and more specifically in access to primary care services, consistent with precarious performance by the SUS in the municipality, but also reflect inadequate sanitation conditions, as described previously. The results highlight the need for monitoring these conditions and an on-going analysis of the issue of health services organization in a municipality that is expected to experience the impact of a petrochemical complex.

Acknowledgments

The authors wish to thank the National School of Public Health, Oswaldo Cruz Foundation and the entire team of the Executive Master's Program in Health Surveillance of the Eastern Region of Rio de Janeiro State for contributing to the exchange of ideas.

References

Publication Dates

History