Hospital readmissions for ambulatory care sensitive conditions in children under 5 years old in the Unified Health System

Abstract

This study aimed to analyze readmissions for Ambulatory Care Sensitive Conditions (ACSC) in children under 5 years in the Unified Health System (SUS) between the years 2009 and 2015, to identify the main cause groups and associated factors. This retrospective cohort study used data from the National Health Database, an individual-centered database constructed from deterministic-probabilistic data matching from the main information systems of SUS. The study population consisted of children under 5 years old hospitalized within the SUS with a primary diagnosis listed in the Brazilian Ambulatory Care Sensitive Conditions. Sociodemographic, clinical, hospitalization-related, and hospital service-related factors were considered. Factors associated with readmission included being an infant (between 28 days and 2 years of age), being male, having been initially hospitalized in a private or non-profit institution, residing in the same municipality as the hospital, and the initial hospitalization cause being related to epilepsy, asthma, nutritional deficiencies, pulmonary diseases, bacterial pneumonia and nose, ear, and throat infections.

Keywords:
Primary Health Care; Child Health; Hospital Readmission; Ambulatory Care Sensitive Conditions

Resumo

O objetivo deste estudo foi analisar as reinternações por Condições Sensíveis à Atenção Primária (CSAP) em menores de 5 anos no Sistema Único de Saúde (SUS) entre os anos de 2009 e 2015, a fim de identificar os principais grupos de causas e fatores associados. Trata-se de um estudo de coorte retrospectiva que utilizou informações da Base Nacional de Saúde, banco de dados centrado no indivíduo e construído a partir de pareamento determinístico-probabilístico de dados dos principais sistemas de informação do SUS. A população do estudo compreendeu crianças menores de 5 anos hospitalizadas no âmbito do SUS com diagnóstico principal registrado na Lista Brasileira de Condições Sensíveis à Atenção Primária. Foram considerados fatores sociodemográficos, clínicos, referentes à hospitalização e aos serviços hospitalares. Os fatores associados à ocorrência da reinternação foram: ser lactente (ter entre 28 dias e 2 anos de idade), ser do sexo masculino, ter sido hospitalizado em instituição privada ou privada sem fins lucrativos na primeira internação, residir no mesmo município da instituição hospitalar e ter tido como causa da internação inicial epilepsias, asma, deficiências nutricionais, doenças pulmonares, pneumonias bacterianas e infecção de nariz, ouvido e garganta.

Palavras-chave:
Atenção Primária à Saúde; Saúde da Criança; Readmissão hospitalar; Condições Sensíveis à Atenção Primária

Resumen

El estudio tuvo como objetivo analizar las readmisiones por Condiciones Sensibles a la Atención Primaria (CSAP) en niños menores de 5 años en el Sistema Único de Salud (SUS) entre 2009 y 2015, con el fin de identificar los principales grupos de causas y factores asociados. Se trata de un estudio de cohorte retrospectivo que utilizó información de la Base Nacional de Salud, una base de datos centrada en el individuo construida a partir del emparejamiento determinista-probabilístico de datos de los principales sistemas de información del SUS. La población del estudio estuvo compuesta por niños menores de 5 años hospitalizados en el ámbito del SUS con diagnóstico principal registrado en la Lista Brasileña de Condiciones Sensibles a la Atención Primaria. Se consideraron factores sociodemográficos y clínicos relacionados con la hospitalización y los servicios hospitalarios. Los factores asociados a la ocurrencia de reingreso fueron: ser lactante (entre 28 días y 2 años de edad), ser hombre, haber estado internado en una institución privada o privada sin fines de lucro durante la primera hospitalización, residir en el mismo municipio que la institución hospitalaria y la causa de la hospitalización inicial fue epilepsia, asma, deficiencias nutricionales, enfermedades pulmonares, neumonía bacteriana e infecciones de nariz, oído y garganta.

Palavras clave:
Atención Primaria de Salud; Salud Infantil; Readmisión hospitalaria; Condiciones Sensibles a la Atención Ambulatoria

Introduction

Children represent one of the population groups that is most vulnerable to the impacts of public health policies, as they are strongly influenced by the physical and social environment in which they live, encountering specific barriers to accessing health care and means of survival. Children are considered to be individuals between the ages of zero and nine, while early childhood is defined as individuals between the ages of zero and five¹. In this light, the profile of morbidity and mortality in childhood, the period of greatest vulnerability to health problems, is a basic parameter for increasing actions and establishing priorities in public policies^1,2. The indicator of hospital admissions due to Ambulatory Care Sensitive Conditions (ACSC) is widely used as a synonym for avoidable hospitalizations, as it encompasses diagnoses for which adequate primary care reduces the risk of hospitalization or its frequency, by preventing the worsening of acute diseases and effectively managing chronic health conditions³. Hospitalization rates due to ACSC can be used to indirectly evaluate barriers in access to specific subgroups, performance, and quality of Primary Health Care (PHC)⁴.

In this sense, the assessment and monitoring of the occurrence of unplanned hospital readmissions through the ACSC indicator presents promising information toward understanding the profile of children most vulnerable to this outcome, as well as the patterns, causes, and frequency of these preventable readmissions.

A readmission is a repeated hospitalization after discharge^5,6, and the time interval considered to characterize a new hospitalization as a readmission depends on the objective with which this event is evaluated. Avoidable readmissions represent a quality metric, since, in addition to generating high costs, they interrupt the routine of the child and their family, expose the patient to the risks of hospitalization, and are frequently associated with adverse health outcomes⁶.

Avoidable readmissions are multifactorial, and therefore the scope of their evaluation should go beyond hospital services, since their occurrence may result from factors unrelated to hospital management, but from the provision of primary care and adequate follow-up after discharge^5,6. In these cases, the evaluation of late readmissions, especially up to 12 months, is the most appropriate period^5,7.

Recent literature still focuses on the study of hospitalizations, and little is known about hospital readmissions due to ACSC, especially at the national level. Exploring this gap in knowledge about readmissions due to preventable conditions in the Unified Health System (SUS) can aid health professionals to better understand the patterns and factors associated with these events, thereby helping to reduce such unfavorable outcomes.

Therefore, the objective of this study was to analyze hospitalizations due to ACSC in children under 5 years of age in SUS, between 2009 and 2015, in order to identify the factors associated with the occurrence of readmissions due to these causes, as well as to outline the sociodemographic and clinical profile of the population subject to them.

Methods

This work is a retrospective cohort study of secondary data from patients selected from the National Health Database of all 26 Brazilian states and the Federal District. The National Health Database consists of an individual-centered database, formulated using deterministic-probabilistic record linkage techniques that integrated data from the main information systems of the Unified Health System (SIA, SIH, SIM) from 2000 to 2015, not including SINAN (2008-2015)⁸.

The study population consists of children under 5 years of age (0 to 4 years, 11 months, and 29 days) hospitalized for ACSC within the scope of SUS. The cases were selected based on the main diagnosis of the Hospital Admission Authorizations (Autorizações de Internação Hospitalar - AIH) belonging to one of the 19 groups of causes of the Brazilian List of ACSC⁹, coded according to the International Statistical Classification of Diseases and Related Health Problems - 10th edition (ICD-10).

The groups of causes belonging to the Brazilian List are: 1. Diseases preventable by immunization and sensitive conditions; 2. Infectious gastroenteritis; 3. Anemia; 4. Nutritional deficiencies; 5. Ear, nose, and throat infections; 6. Bacterial pneumonia; 7. Asthma; 8. Lung diseases; 9. Hypertension; 10. Angina; 11. Heart failure; 12. Cerebrovascular diseases; 13. Diabetes Mellitus; 14. Epilepsy; 15. Kidney and urinary tract infections; 16. Infection of the skin and subcutaneous tissue; 17. Inflammatory disease of the female pelvic organs; 18. Gastrointestinal ulcer; and 19. Diseases related to prenatal care and childbirth.

All 19 groups were maintained in both the descriptive analysis and the statistical model. However, the nine most frequent groups of causes of hospitalization were listed one by one, namely: 2. Infectious gastroenteritis, 6. Bacterial pneumonia, 7. Asthma, 8. Lung diseases, 15. Kidney and urinary tract infections, 16. Skin and subcutaneous tissue infections, 5. Ear, nose, and throat infections, 14. Epilepsy and 4. Nutritional deficiencies. The remaining ten groups of causes each presented less than 1% of involvement in the population monitored, and for this reason they were aggregated in the analysis subcategory of “Others”. The inclusion criteria were children, aged 0 to 4 years, 11 months, and 29 days, of both sexes, with hospitalization due to ACSC through SUS between July 1, 2009 and June 30, 2014. The period of entry into the cohort ended one year before the end of the readmission monitoring period to allow for the occurrence of any readmission within 365 days after the discharge of children who entered the last year of the cohort.

The period of one year before the start of the cohort (July 2008 to June 2009) was examined for a hospitalization due to ACSC for those children who had their first hospitalization due to ACSC in the first year of the cohort (between July 2009 and June 2010). Based on the results of this search, children whose only hospitalization for ACSC occurred between July 2009 and June 2010, but who had been hospitalized for ACSC prior to their entry into the cohort (between July 2008 and June 2009), were excluded, as the hospitalization that led to their entry into the cohort was, in fact, a readmission.

Pediatric readmissions due to ACSC occurring within 365 days after discharge from an index hospitalization due to ACSC were considered potentially preventable. The study cohort included the possibility of more than one index hospitalization for the same child if the period considered for readmission was exceeded.

The outcome variable was readmission due to ACSC within 365 days after an index hospitalization due to ACSC, categorized as “No readmission” or “Readmission”.

The explanatory variables were linked to sociodemographic factors of the study population, clinical factors, and the hospital establishment. The sociodemographic variables included age at first hospitalization (in years), sex (female, male), race/color (white, black, brown, Asian, indigenous, unknown), region of residence (North, Northeast, Southeast, South, Midwest), residing in the same municipality as the hospital (no, yes), and type of municipality of residence (rural, intermediate, and urban). To understand the outcome of readmissions in each phase of childhood, age was divided into three subgroups for analysis, considering the specificities of each stratum: neonates (0 to 27 days), infants (28 days to 2 years), and preschoolers (2 to 4 years, 11 months and 29 days).

The clinical variables included the primary diagnosis at admission, occurrence of death during the period (no, yes), and number of hospitalizations per patient. Finally, the variable related to the hospitalization facility funded by SUS is the legal nature of the facility (public, private, private non-profit).

The bivariate analysis of the association between explanatory variables and the outcome included Pearson’s Chi-square test and binary logistic regression, from which all variables with a p-value of up to 0.20 were selected to be part of the multiple logistic regression model.

The variable of municipality typology (rural, urban, and intermediate) was removed from the final statistical model because it presented multicollinearity identified by the value of the Variance Inflation Factor (VIF). Additionally, the race/color variable could not be included in the statistical model due to the weakness in the completeness of the information (30% of missing data) and was only maintained in the descriptive analysis.

Although there were children with more than one index hospitalization, these represented only 3.16% of the individuals and the adjustment of a generalized linear model with random effects for the individuals led, in practice, to the same results as the binary logistic regression model. Therefore, the simplicity of the latter model was chosen. In all statistical analyses, a significance level of 5% was adopted for hypothesis testing, along with a 95% confidence interval.

The database used in this study was extracted from the National Health Database via Structured Query Language (SQL) and analyzed in the R Project for Statistical Computing, version 4.1.1, with the help of its packages.

This research was approved by the Research Ethics Committee of the Federal University of Minas Gerais (UFMG) through the Certificate of Presentation of Ethical Appreciation (CAAE) No. 44121315.2.0000.5149.

Results

A total of 1,559,880 children under 5 years of age, who were hospitalized for some type of ACSC during the six-year cohort period, were selected, totaling 1,939,788 hospitalizations. Of the total number of children selected, 194,788 (12.5%) had readmissions that were considered avoidable.

The majority of children hospitalized for ACSC during the period were male, with a higher incidence among children who were hospitalized (Table 1). The geographic region with the highest number of patients hospitalized and readmitted was the Northeast, followed by the Southeast, North, and South, while the Midwest region had the lowest values. The North and Northeast regions had the highest incidence of infectious gastroenteritis, while the South and Southeast regions had a predominance of lung diseases. Bacterial pneumonias were the most frequent group of causes in the Midwest and Southeast regions.

Table 1
Characteristics of patients under 5 years of age hospitalized for Ambulatory Care Sensitive Conditions (ACSC) by SUS, 2009 to 2015, Brazil.

Characteristics	Patients n= 1,559,880 (100%)	Readmission n= 194,788 (12.5%)
Age
Neonates	46,521 (3.0%)	4,699 (2.4%)
Infants	953,179 (61.1%)	145,023 (74.5%)
Preschoolers	560,180 (35.9%)	45,066 (23.1%)
Sex
Female	723,811 (46.4%)	86,556 (44.4%)
Male	836,069 (53.6%)	108,232 (55.6%)
Region
North	244,151 (15.7%)	29,638 (15.2%)
Northeast	536,268 (34.4%)	66,104 (33.9%)
Southeast	464,180 (29.8%)	58,641 (30.1%)
South	189,174 (12.1%)	25,781 (13.2%)
Midwest	126,107 (8.1%)	14,624 (7.51%)
Race/skin color
White	395,661 (25.4%)	52,833 (27.1%)
Black	28,870 (1.8%)	3,792 (1.9%)
Brown	647,148 (41.5%)	96,239 (49.4%)
Yellow	6,094 (0.4%)	920 (0,5%)
Indigenous	13,272 (0.9%)	2,649 (1.4%)
Ignored	469,835 (30.1%)	38,355 (19.7%)
Lives in the municipality of readmission
Non-resident	268,472 (17.2%)	34,433 (17.7%)
Resident	1,291,408 (82.8%)	160,355 (82.3%)
CSAP Group
2 - Infectious gastroenteritis	624,683 (40.0%)	65,903 (33.8%)
6 - Bacterial pneumonia	215,796 (13.8%)	31,424 (16.1%)
8 - Lung diseases	213,034 (13.7%)	33,972 (17.4%)
7 - Asthma	206,947 (13.3%)	30,525 (15.7%)
15 - Kidney and Urinary Tract Infection	86,578 (5.6%)	9,033 (4.7%)
16 - Infection of the skin and subcutaneous tissue	67,418 (4.3%)	3,961 (2.0%)
5 - Nose, ear and throat infections	51,226 (3.3%)	5,352 (2.8%)
14 - Epilepsies	40,222 (2.6%)	7,221(3.7%)
4 - Nutritional Deficiencies	15,105 (1.0%)	2,346 (1.2%)
Others	38,871 (2.4%)	5,051 (2.6%)
Death
No	1,540,558 (98.8%)	190,400 (97.7%)
Yes	19,322 (1.2%)	4,388 (2.3%)
Number of readmissions
One	-	145,471 (74,.7%)
Two	-	31,568 (16.2%)
Three	-	9,997 (5.1 Idade%)
Four or more	-	7,752 (4.0%)
Readmission by the same CSAP group
Yes	-	112,504 (57.8%)
No	-	82,284 (42.2%)
Number of readmissions by the same CSAP group
One	-	91,805 (81.6%)
Two	-	14,648 (13.0%)
Three	-	3,622 (3.2%)
Four or more	-	2,429 (2.2%)

• Source: National Individual-Centered Health Database (Base Nacional de Saúde Centrada no Indivíduo), 2023.

A higher incidence of hospitalized patients was found mainly in the infant age group, especially among readmitted children. This age group, from 28 days to 2 years, represented 61% of hospitalized children and 74.3% of children with readmission.

The most frequent group of causes of hospitalization and readmission was infectious gastroenteritis and its complications, with the most frequent ICD-10 diagnosis in this group being A09 - Diarrhea and infectious gastroenteritis of presumed origin. The other groups of causes with the highest occurrence were, respectively, bacterial pneumonia, lung diseases, asthma, and kidney and urinary tract infections. In general, the proportion of diseases affecting the respiratory system (bacterial pneumonia, lung diseases, and asthma) is higher in readmitted children, when compared to children who were not readmitted, and, when aggregated, exceeds the proportion of infectious gastroenteritis (49.2% and 33.8%, respectively).

The five main groups of causes for patients with “no readmission” and readmission remained the same. However, for readmission children, lung diseases were the second most frequent group of causes, and bacterial pneumonias represented the third. Furthermore, skin and subcutaneous tissue infections were the sixth leading cause of hospitalizations, while among readmissions, epilepsy represented the sixth most frequent cause.

Among children with one or more readmissions, 2.3% died during the follow-up period, a proportion twice as high as among children who were not readmitted (1.1%). This same pattern was maintained for deaths occurring during hospitalization (0.4% and 0.7%, respectively).

Most readmission patients had a single readmission during the cohort period (74.7%), both for the same ACSC group (81.6%) and for different groups of causes. The cause of readmissions was mostly due to diagnoses belonging to the same ACSC group as the index hospitalization (57.8%).

Despite its relevance for identifying the profile of children who had preventable hospitalizations, the race/color variable presented a significant deficit in its completeness, in which 30.1% of the patients hospitalized during this period had this information ignored when filling out the AIH. This fact imposes a difficulty in the sociodemographic assessment and health status of specific population subgroups. In readmitted children, a lower percentage of ignored responses for this variable was observed, reaching 19.7%. In this scenario, the brown race/color (41.5%) was that which presented the highest occurrence of patients hospitalized for ACSC, followed by the white race/color (25.4%), similar to what occurred for patients with readmission, respectively 49.4% and 27.1%. Children of indigenous race/color also presented a higher occurrence in the readmission group, representing a proportional increase of 82% in relation to the non-readmitted children. Table 2 presents the results of the adjustment of the simple and multiple binary logistic regression models. In line with the descriptive results, the analysis of factors associated with readmission due to ACSC identified that age was the covariate that presented the greatest magnitude of association with the occurrence of this event: children aged 28 days to 2 years had a 100% greater chance of readmission due to ACSC than did preschool children aged 2 to 4 years (OR=2.00). Neonates had a 27% greater chance of readmission when compared to preschool children.

Table 2
Odds ratios for hospital readmission due to Ambulatory Care Sensitive Conditions (ACSC) in children under 5 years of age, 2009 to 2015, in SUS - Brazil.

	Individual models		Multiple models
Variables	OR (95% CI)	p-value	OR (95% CI)	p-value
Age group
Preschoolers	1		1
Infants	2.05 (2.03-2.07)	< 0.001	2.00 (1.98-2.02)	< 0.001
Neonates	1.29 (1.25-1.33)	< 0.001	1.30 (1.23-1.31)	< 0.001
Sex
Female	1		1
Male	1.09 (1.08-1.11)	< 0.001	1.06 (1.05-1.07)	< 0.001
Region
Southeast	1		1
Northeast	0.97 (0.96-0.99)	< 0.001	1.10 (1.09-1.12)	< 0.001
North	0.96 (1.08-1.11)	< 0.001	1.07 (1.06-1.09)	< 0.001
South	1.09 (1.08-1.11)	< 0.001	1.04 (1.02-1.05)	< 0.001
Midwest	0.91 (0.89-0.93)	0.003	0.97 (0.95-0.99)	0.003
Residence in the same municipality as the hospital
Non-resident	1		1
Resident	0.96 (0.95-0.98)	< 0.001	1.02 (1.01-1.04)	< 0.001
Legal nature of the hospital
Public	1		1
Private	1.13 (1.11-1.14)	< 0.001	1.16 (1.15-1.18)	< 0.001
Private Non-profit	1.14 (1.12-1.15)	< 0.001	1.12 (1.11-1.13)	< 0.001
Reason for initial admission
2 - Infectious gastroenteritis	1		1
6 - Bacterial pneumonia	1.45 (1.42-1.47)	< 0.001	1.41 (1.39-1.43)	< 0.001
8 - Lung diseases	1.61 (1.58-1.63)	< 0.001	1.43 (1.41-1.46)	< 0.001
7 - Asthma	1.47 (1.44-1.49)	< 0.001	1.54 (1.52-1.56)	< 0.001
15 - Kidney and urinary tract infections	0.98 (0.96-1.01)	0.459	1.01 (0.98-1.03)	0.459
16 - Skin and subcutaneous tissue infections	0.53 (0.51-0.55)	< 0.001	0.59 (0.57-0.61)	< 0.001
5 - Nose, ear, and throat infections	0.99 (0.96-1.02)	0.004	1.04 (1.01-1.08)	0.004
14 - Epilepsy	1.86 (1.81-1.91)	< 0.001	1.96 (1.91-2.01)	< 0.001
4 - Nutritional deficiencies	1.56 (1.49-1.63)	< 0.001	1.50 (1.43-1.56)	< 0.001
Others	1.27 (1.23-1.31)	< 0.001	1.31 (1.27-1.35)	< 0.001

• Source: National Individual-Centered Health Database (Base Nacional de Saúde Centrada no Indivíduo), 2023.

Sex was also an important factor in the outcome of readmission, as male children had a 6% higher chance of readmission due to ACSC than did female children. Regarding the child’s place of residence, those living in the same municipality as the hospital had a 2% higher chance of readmission due to ACSC.

Regarding the Southeast region, which has the largest population, children living in the Northeast region had a 10% higher chance of readmission due to ACSC. Living in the North and South regions also increased the chances of readmission by 7% and 4%, respectively, while living in the Midwest region presented a protective factor for readmission (OR = 0.97).

Infectious gastroenteritis was the main overall cause of hospitalization and readmission due to ACSC, as shown in the descriptive results, and were therefore used as a reference to calculate the magnitude of association of the other causes with the occurrence of readmission.

Regarding the diagnosis of the index hospitalization, epilepsy was the group of causes most strongly associated with readmission: children hospitalized for epilepsy had a 96% greater chance of readmission than did those hospitalized for gastroenteritis. Regarding diseases affecting the respiratory tract, children who were hospitalized with lung diseases (OR=1.43) or bacterial pneumonia (OR=1.41) had similar chances of readmission when compared to those hospitalized with gastroenteritis, while the diagnosis of asthma in the index hospitalization, as it is a chronic disease, increased the chance of readmission (OR=1.54) in relation to gastroenteritis.

The diagnosis of the first hospitalization due to nutritional deficiencies was also a risk factor for readmission, with a 50% greater chance of the outcome’s occurrence. Similarly, a diagnosis in one of the other 10 groups of causes on the Brazilian list of ACSCs proved to be a risk factor for readmission (OR=1.31) in relation to gastroenteritis.

An index hospitalization for skin and subcutaneous tissue infection represented the only diagnosis that was a protective factor for readmission, with children hospitalized for this cause showing a 41% lower chance of readmission than did children hospitalized for gastroenteritis.

The legal nature of the hospital where the index hospitalization occurred also generated differences in the chances of readmission for ACSCs. Using public hospitals as a reference, which provided 50.5% of the care for hospitalized children, children hospitalized in private or non-profit institutions showed a higher chance of readmission for preventable causes after discharge (16% and 12%, respectively).

Discussion

This study, conducted using population data on readmissions due to ACSC in children under 5 years of age in SUS, shows that the factors associated with the occurrence of readmission were: being an infant (between 28 days and 2 years of age); being male; having been hospitalized in a private or non-profit institution during the first admission; residing in the same municipality as the hospital institution and having had epilepsy; asthma; nutritional deficiencies; lung diseases; bacterial pneumonia; and nose, ear, and throat infections as the cause of the initial admission.

Many factors may be related to the greater chance of readmission due to ACSC in newborns and infants. Physiologically, children are more vulnerable to a wide range of diseases and injuries in this age group, as they are in a period of adaptation to the external environment and of development and maturation of the organic systems, especially the immune, metabolic, and pulmonary systems^10,11,12. Conditions related to pregnancy, childbirth, and postpartum can be contributing factors to the vulnerability of children’s health during this period. Prematurity generates a higher risk of morbidity and mortality in the short term, and can result in problems with growth and development in the long term^13,14, predisposing a greater need for hospital monitoring due to the severity of the health condition in the event of illness.

Even though newborns are more likely to be readmitted than preschoolers, being an infant is a risk factor of greater magnitude for the outcome. This fact can be explained by the short time period that characterizes the neonatal age group and their greater health fragility, that is, once newborns are hospitalized, they tend to have a longer hospital stay, leaving a short time period for new readmissions within the first 27 days of life.

A cross-sectional study that evaluated breastfeeding in children under 2 years of age found that breastfeeding, whether exclusive or not, reduced the occurrence of prevalent diseases in childhood, which include ACSC¹⁵.

In addition to the lack of breastfeeding, malnutrition and diarrhea are associated with the family’s living conditions, such as insufficient income, low level of education, a lack of basic sanitation, and barriers in access to health services, which are also factors of susceptibility to illness and death in this age group¹¹. Studies carried out on hospital admissions due to ACSC have already demonstrated a greater susceptibility of children under one and two years of age to the outcome of hospitalization^10,16,17,18. Thus, the present study highlights the role of monitoring and surveillance of PHC to mitigate these unfavorable outcomes¹⁹.

Male children were more likely to be readmitted due to ACSC than were female children. Numerous studies, both on hospitalizations due to ACSC and unplanned pediatric readmissions, have found that boys are more susceptible to hospitalizations than girls^{16,17,20,21,22}.

In general, infant mortality is also higher in male children. The causes of these differences are related to biological factors, which indicate a greater vulnerability of male children to some types of diseases linked to external causes, such as diarrhea, hemorrhages, and pneumonia, some of the main diagnoses for hospitalizations due to ACSC. Furthermore, male fetuses have a higher risk of miscarriage due to the higher incidence of genetic alterations¹¹.

Children who live in the same municipality as the hospital institution where they are admitted have a higher chance of being readmitted, showing that the availability of hospital services influences families’ search for these services. Caminal-Homar and Casanova-Matutano²³, when carrying out a theoretical evaluation of the functioning of PHC and its relationship with hospitalizations due to ACSC, indicate that the ease of access to specialized services, such as hospitals and emergency services, leads to their inadequate and excessive use.

Despite the scarcity of observational studies evaluating this characteristic, in Italy, children hospitalized for preventable causes were more likely to live in the same province as the hospital²⁴. Another study conducted in Rio de Janeiro found that the availability of hospital beds influences the occurrence of hospitalizations for ACSC²⁵, but spatial accessibility is not always significantly associated with the rate of use of health services²⁶.

The main causes of readmission are in line with the Brazilian epidemiological situation, which involves the concomitant presence of infectious and deficient diseases, external causes, and chronic diseases²⁷. The results showed that patients were more commonly readmitted for the same diagnosis as the initial hospitalization, which also occurred in other studies^28,29. In addition, the cause of the index hospitalization was associated with the occurrence of subsequent readmission.

Gastroenteritis was the main cause of hospitalization and readmission for ACSC. This is still considered a public health problem worldwide, especially in developing countries and among children under 5 years of age¹⁸. In Brazil, environmental sanitation conditions, despite having improved in recent years, are still deficient. Studies indicate that there is a proven association between inadequate sanitation and cases of gastroenteritis. Living in a household with sewage coverage provided by a general network is inversely related to hospitalizations due to waterborne illnesses, such as diarrheal diseases^30,31.

A lack of access to treated water, low family income, and low maternal education also influence the occurrence of gastroenteritis. Timely use of oral rehydration therapy, improvements in water quality, basic sanitation, and hygiene conditions of the population could reduce the cases of these diseases^18,30,31.

Epilepsy and asthma were the diagnoses with the greatest magnitude of association with the outcome of readmission. Both are chronic diseases, which depend on adequate monitoring in health services, among other factors. A study that evaluated hospitalizations due to ACSC in the pediatric population in Rondônia (BR) identified an annual increasing trend between 2008 and 2019 in hospitalizations due to epilepsy in children, aged 1 to 4 years³². Regarding childhood asthma, hospital readmission may represent the worsening of the disease with more frequent exacerbations. This fact may be due to low adherence to treatment or limited access to monitoring in PHC³³.

The timing and causes of pediatric readmissions are strongly influenced by the initial diagnosis. Therefore, strategies for assessing and reducing readmissions need to take into account the index diagnosis to better identify and mitigate the factors associated with the causes of readmission. It is also worth noting that the differences found between the three age groups evaluated in this study also extend to variations in the disease profile and differences in the main groups of causes of readmission, both of which should be taken into consideration.

The type of hospital management impacts the level of administrative autonomy, influencing factors, such as the ability to introduce care protocols, cost management, and organization of care. Thus, the method of contracting and payment defines the incentive structure under which providers will provide care. Particularly in the Brazilian private sector, payment per procedure predominates, while in public hospitals there is a greater diversity of remuneration structures. When payment per procedure predominates, there are strong incentives to increase production, especially in small and medium-sized hospitals that are more dependent on these resources³⁴.

This is a challenge for hospital management in SUS, since Brazil’s geopolitical configuration is marked by small municipalities and low population scale, requiring very diverse institutional arrangements that include multiple interactions between public and private providers to guarantee access to hospital beds³⁴.

In the multivariate analysis, in relation to the Southeast region, children who lived in the Northeast and North regions had a 10% and 7% greater chance, respectively, of readmission due to ACSC. The country’s macro-regions vary in terms of the main causes of hospital readmissions. These differences may be related to climate issues in the different regions, as well as infrastructure issues, such as the quality of and access to basic sanitation.

Historically, the Northeast and North regions have had the highest incidence of hospitalizations due to ACSC in children³⁵. In addition, both regions remain below the national average in terms of water supply, sewage, and garbage collection³⁶. Thus, despite not having the largest population, the North and Northeast regions have the highest incidence and risk factors for avoidable hospitalizations, since infectious causes are directly related to environmental and socioeconomic issues^30,31. Increasing actions that mitigate regional inequalities and rigorously monitor these children who are most susceptible to avoidable readmissions is essential to reducing this outcome.

The limitations of this study include the failure to assess the presence of multimorbidity and pre-existing diseases, such as complex chronic conditions, which have proven to increase the risk of readmissions. Other limitations include a possible information bias, since only secondary data were used, and the incompleteness of the information on the race/color variable, which did not allow it to be included in the statistical model.

It is expected that the research carried out in this study can contribute to the qualification of information about the health profile of children under five years of age who use SUS and mitigate/monitor the factors that may culminate in this outcome.

Likewise, it is believed that these findings can help professionals, managers. and policymakers, especially in PHC and hospitals, to analyze their flows, identifying the target conditions and characteristics that may predispose a child to readmission. Finally, it is suggested that new contemporary studies that address this topic should be carried out to enrich the existing literature and better evaluate the population subgroups and characteristics not included in this study.

Conclusion

The findings on avoidable hospital readmissions were unprecedented in the SUS setting. The clinical profile of readmitted patients was predominantly male, brown-skinned, infants (28 days to 2 years of age), who lived in the same municipality as the hospital and became ill with infectious gastroenteritis, lung disease, and bacterial pneumonia. These children were also predominantly from the Northeast region and lived in urban areas. Most of the care was provided in public hospitals, on an emergency basis. Most of the children hospitalized for ACSC were not readmitted, and of those who were readmitted, most were only readmitted once during the study period.

References

• ¹ Brasil. Ministério da Saúde (MS). Política Nacional de Atenção Integral à Saúde da Criança: orientações para implementação. Brasília, DF; 2018 [acessado 2021 Nov. 25].
• ² Bugelli A, Borgès da Silva R, Dowbor L, Sicotte C. The Determinants of Infant Mortality in Brazil, 2010-2020: A Scoping Review. Int J Environ Res Public Health 2021; 18(12):6464.
• ³ Billings J, Zeitel L, Lukomnik J, Carey TS, Blank AE, Newman L. Impact of socioeconomic status on hospital use in New York City. Health Aff 1993; 12(1): 162-173.
• ⁴ Billings J, Anderson GM, Newman LS. Recent findings on preventable hospitalizations. Health Aff 1996; 15(3):239-249.
• ⁵ Benbassat J, Taragin M. Hospital Readmissions as a Measure of Quality of Health Care: Advantages and Limitations. Arch Intern Med 2000; 160(8):1074-1081.
• ⁶ Kneepkens EL, Brouwers C, Singotani RG, Bruijne MC, Karapinar-Çarkit F. How do studies assess the preventability of readmissions? A systematic review with narrative synthesis. BMC Med Res Methodol 2019; 19(1):128.
• ⁷ Shimizu E, Glaspy K, Witt MD, Poon K, Black S, Schwartz S, Bholat T, Diaz N, Kuo A, Spellberg B. Readmissions at a Public Safety Net Hospital. PLoS ONE 2014; 9(3):e91244.
• ⁸ Guerra Junior AA, Acurcio FA, Reis IA, Santos N, Ávila J, Dias LV, Santos N, Reis A, Acurcio FA, Meira Junior, W. Building the National Database of Health Centred on the Individual: Administrative and Epidemiological Record Linkage, Brazil, 2000-2015. Int J Popul Data Sci 2018; 3(1). Doi:https://doi.org/10.23889/ijpds.v3i1.446
  » https://doi.org/10.23889/ijpds.v3i1.446
• ⁹ Brasil. Ministério da Saúde (MS). Portaria nº 221, de 17 de abril de 2008. Diário Oficial da União 2008.
• ¹⁰ Costa LQ, Pinto Junior EP, Silva MGC. Time trends in hospitalizations for Ambulatory Care Sensitive Conditions among children under five years old in Ceará, Brazil, 2000-2012. Epidemiol Serv Saude 2017; 26(1):51-60.
• ¹¹ Alves TF, Coelho AB. Mortalidade infantil e gênero no Brasil: uma investigação usando dados em painel. Cienc Saúde Coletiva 2021; 26(4):1259-1264.
• ¹² Munhoz TN, Santos IS, Blumenberg C, Barcelos RS, Bortolotto CC, Matijasevich A, Santos Júnior HG, Santos LM, Correa LL, Souza MR, Lira PIC, Altafim ERP, Macana EC, Victora CG. Fatores associados ao desenvolvimento infantil em crianças brasileiras: linha de base da avaliação do impacto do Programa Criança Feliz. Cad Saude Publica 2020; 38(2): e00316920.
• ¹³ França EB, Lansky S, Rego MAS, Malta DC, França JS, Teixeira R, Porto D, Almeida MF, Souza MFM, Szwarcwald CL, Mooney M, Naghavi M, Vasconcelos AMN. Principais causas da mortalidade na infância no Brasil, em 1990 e 2015: estimativas do estudo de Carga Global de Doença. Rev Bras Epidemiol 2017; 20(1):46-60.
• ¹⁴ Barros FC, Rabelo Neto DL, Villar J, Kennedy S, Silveira MF, Diaz- Rosello JL, Victora CG. Caesarean sections and the prevalence of preterm and early-term births in Brazil: secondary analyses of national birth registration. BMJ Open 2018; 8(8):e021538.
• ¹⁵ Nass EMA, Marcon SS, Teston EF, Leal LP, Ichisato SMT, Toso BRGDO, Moreira MAR, Bernardino FBS. Amamentação e as doenças prevalentes nos primeiros dois anos de vida da criança: estudo transversal. Rev Bras Enferm 2022; 75(6):e20210534.
• ¹⁶ Ribeiro MGC, Araujo Filho ACA, Rocha SS. Children's hospitalizations by sensitive conditions in primary care in the Northeast of Brazil. Rev Bras Saúde Mater Infant 2019; 19(2):491-498.
• ¹⁷ Amaral JV, Araújo Filho ACA, Rocha SS. Hospitalizações infantis por condições sensíveis à atenção primária em cidade brasileira. Av enferm 2020; 38(1):46-54.
• ¹⁸ Konstantyner T, Mais LA, Taddei JAAC. Factors associated with avoidable hospitalisation of children younger than 2 years old: the 2006 Brazilian National Demographic Health Survey. Int J Equity Health 2015; 14:69.
• ¹⁹ Pasklan ANP, Rocha TAH, Queiroz RCS, Rocha NCS, Facchini LA, Thomaz EBAF. Are Primary Health Care Features Associated with Reduced Late Neonatal Mortality in Brazil? An Ecological Study. Matern Child Health J 2021; 26:1790-1799.
• ²⁰ Araujo EMN, Costa GMC, Pedraza DF. Hospitalizations due to primary care-sensitive conditions among children under five years of age: cross-sectional study. São Paulo Med J 2017; 135(3):270-276.
• ²¹ Mariano TSO, Nedel FB. Hospitalização por Condições Sensíveis à Atenção Primária em menores de cinco anos de idade em Santa Catarina, 2012: estudo descritivo. Epidem Serv Saúde 2018; 27(3):e2017322.
• ²² Brown CM, Williams DJ, Hall M, Freundlich KL, Johnson DP, Lind C, Rehm K, Frost PA, Doupnik SK, Ibrahim D, Patrick S, Howard LM, Gay JC. Trends in Length of Stay and Readmissions in Children's Hospitals. Hosp Pediatr 2021; 11(6):554-562.
• ²³ Caminal Homar J, Casanova Matutano C. La evaluación de la atención primaria y las hospitalizaciones por ambulatory care sensitive conditions. Marco conceptual. Atención Primaria 2003; 31(1):61-5.
• ²⁴ Zucco R, Pileggi C, Vancheri M, Papadopoli R, Nobile CGA, Pavia M. Preventable pediatric hospitalizations and access to primary health care in Italy. PloS One 2019; 14(10):e0221852.
• ²⁵ Botelho JF, Portela MC. Risk of misinterpretation of trends in hospital admissions for primary care sensitive conditions in local contexts: Itaboraí, Rio de Janeiro State, Brazil, 2006-2011. Cad Saúde Pública 2017; 33(3):e00050915.
• ²⁶ Mudd AE, Michael YL, Diez-Roux AV, Maltenfort M, Moore K, Melly S, Lê-Scherban F, Forrest CB. Primary Care Accessibility Effects on Health Care Utilization Among Urban Children. Acad Pediatr 2020; 20(6):
• ²⁷ Camelo MS, Rehem TCMSB. Internações por condições sensíveis à atenção primária em pediatria no distrito federal: um estudo ecológico exploratório. REME Rev Min Enferm 2019; 23:e-1269.
• ²⁸ Berry JG, Toomey SL, Zaslavsky AM, Jha AK, Nakamura MM, Klein DJ, Feng JY, Shulman S, Chiang VW, Kaplan W, Hall M, Schuster MA. Pediatric Readmission Prevalence and Variability Across Hospitals. JAMA 2013; 309(4):372-380.
• ²⁹ Bucholz EM, Gay JC, Hall M, Harris M, Berry JG. Timing and Causes of Common Pediatric Readmissions. J Pediatrics 2018; 200:240-248.e1.
• ³⁰ Rasella, D. Impacto do Programa Água para Todos (PAT) sobre a morbi-mortalidade por diarreia em crianças do Estado da Bahia. Cad Saude Publica 2013; 19(1):40-50.
• ³¹ Paiva RPS, Souza MFP. Associação entre condições socioeconômicas, sanitárias e de atenção básica e a morbidade hospitalar por doenças de veiculação hídrica no Brasil. Cad Saude Publica 2018; 34(1):e00017316.
• ³² Santos ADS, Castro LR, Freitas JLG, Cavalcante DFB, Pereira PPDS, Oliveira TMCD, Alves JC. Internações por condições sensíveis à atenção primária em crianças, Rondônia, Brasil, 2008-2019. Ciênc saúde coletiva 2023; 28(4):1003-10.
• ³³ Ege MJ. Structural racism and readmission for childhood asthma a quest for causality. J Allergy Clin Immunol 2021; 148(5):1165-1166.
• ³⁴ Botega LA, Andrade MV, Guedes GR. Perfil dos hospitais gerais do Sistema Único de Saúde. Rev Saúde Pública 2020; 54:81.
• ³⁵ Moura BLA, Cunha RC da, Aquino R, Medina MG, Mota ELA, Macinko J, Dourado I. Principais causas de internação por condições sensíveis à atenção primária no Brasil: uma análise por faixa etária e região. Rev Bras Saude Mater Infant 2010; 10:s83-91.
• ³⁶ Massa KHC, Chiavegatto Filho ADP. Saneamento básico e saúde autoavaliada nas capitais brasileiras: uma análise multinível. Rev bras epidemiol 2020; 23:e200050.

• Editors-in-chief:
  Maria Cecília de Souza Minayo, Romeu Gomes, Antônio Augusto Moura da Silva

Publication Dates

• Publication in this collection
  02 June 2025
• Date of issue
  2025

History

• Received
  07 Nov 2023
• Accepted
  09 Apr 2024
• Published
  11 Apr 2024