Artigos

Conteúdo (article):

The influence of climatic conditions on hospital admissions for asthma in children and adolescents living in Belo Horizonte, Brazil – Projeto BH - Viva

Autores:
Cláudia Silva Dias
Instituição:Pontifícia Universidade Católica de Minas Gerais
E-mail:claudiadiaspuc@yahoo.com.br
ORCID: https://orcid.org/0000-0002-1205-8862

Sueli Aparecida Mingoti
Universidade Federal de Minas Gerais - Departamento de Estatística, Instituto de Ciências Exatas
suelimngt@gmail.com
ORCID: https://orcid.org/0000-0003-3416-4014

Ana Paula Romanelli Ceolin
Universidade Federal de Minas Gerais – Faculdade de Medicina – Observatório de Saúde Urbana
aprc13@gmail.com
ORCID: https://orcid.org/0000-0002-9237-0804

Maria Angélica de Salles Dias - Universidade Federal de Minas Gerais – Faculdade de Medicina – Observatório de Saúde Urbana
angelica@pbh.gov.br
ORCID:https://orcid.org/0000-0002-1891-0585

Amélia Augusta de Lima Friche
Universidade Federal de Minas Gerais – Faculdade de Medicina – Observatório de Saúde Urbana
gutafriche@gmail.com
Orcid: https://orcid.org/0000-0002-2463-0539

Waleska Teixeira Caiaffa
Universidade Federal de Minas Gerais – Faculdade de Medicina
wcaiaffa@medicina.ufmg.br ou caiaffa.waleska@gmail.com
OCID: https://orcid.org/0000-0001-5043-4980

Resumo
Introdução: o risco de hospitalização por asma influenciado pelo disparo das condições climáticas é pouco explorado em Minas Gerais. Objetivos: a) avaliar a influência dos fatores climáticos nas hospitalizações por asma e por infecções virais do trato respiratório inferior (IVTRI), de 2002 a 2012, em crianças e adolescentes residentes em Belo Horizonte (BH) e estimar períodos epidêmicos para as hospitalizações por asma; b) comparar o padrão sazonal local das hospitalizações por asma e IVTRI. Métodos: Utilizando as hospitalizações por asma estratificadas e por bronchiolitis de 0-4 anos, a partir das guias de Internação Hospitalar, do Projeto BH-Viva, modelos estatísticos de regressão foram aplicados para avaliar o relacionamento entre as variáveis. Para estimar períodos epidêmicos foi utilizado o modelo de séries temporais da classe ARIMA. Resultados: Foi observado um incremento nas hospitalizações por asma com aumento da umidade relativa no período pós-chuvas, as hospitalizações por bronchiolitis se associaram a baixos níveis de temperatura máxima e precipitação. Conclusão: períodos mais chuvosos podem propiciar o aumento da umidade dentro e fora do domicílio favorecendo proliferação fúngica. Já os períodos mais frios podem favorecer o aumento da disseminação de vírus.
Palavras-chave: Hospitalização por asma. Clima. Hospitalização por bronquiolite. Criança e adolescente. Saúde Urbana.

Abstract
Introduction: limited research exists on the influence of climatic conditions on the risk of hospital admission for asthma in Minas Gerais, Brazil. Objectives: a) to evaluate the influence of climatic conditions on hospital admissions for asthma and lower respiratory tract infections (LRTIs) among children and adolescents living in Belo Horizonte during the period 2002 to 2012 and identify epidemic peaks of admissions for asthma; b) to compare local seasonal patterns of admissions for asthma and LRTIs. Methods: using hospital admission data from the Projeto BH-Viva stratified by aged group, regression analysis was performed to determine the relationship between the variables. Epidemic peaks were identified using an ARIMA model. Results: There was an increase in admissions for asthma with an increase in relative humidity after rainy periods; admissions for bronchiolitis were associated with low levels of maximum temperature and rainfall. Conclusion: Rainy periods can lead to an increase in indoor and outdoor humidity, facilitating fungal proliferation, while cold periods can lead to an increase in the spread of viruses.
Key words: Hospital admissions for asthma. Climate. Hospital admissions for bronchiolitis. Child and adolescent.

INTRODUCTION
Asthma is a multifactorial disorder characterized as a chronic inflammatory disease with bronchial hyperresponsiveness and variable airflow limitation that is often reversible, either spontaneously or through the use of bronchodilators1,2.
In Brazil, despite a fall in the number of hospital admissions, asthma is the third biggest driver of healthcare costs in Brazil’s national health system3. While studies have shown a decline in admissions for asthma in regions and large cities such as Minas Gerais4 and Belo Horizonte5, this reduction has been less pronounced among people living in areas with a high level of social vulnerability, such as slums6, resulting in asthma inequalities.
According to the National Heart, Lung, and Blood Institute, various interacting individual, lifestyle, and environmental factors may cause asthma exacerbations, including lower respiratory tract infections (LRTIs), contact with indoor and outdoor allergens, and weather changes2,7,8. Moreover, these factors may be influenced by socioeconomic status2.
Respiratory viruses have been recognized as a major factor in the sharp rise of wheezing episodes and asthma exacerbations, especially among children under two years of age. Respiratory syncytial virus is known to be one of the main causes of wheezing episodes in these children9.
Another important factor that can trigger asthma attacks is climatic conditions. Studies conducted in different regions of the world, including Brazil, have shown a link between admissions for asthma and seasonal variations in climate10,11,12. However, the role played by climatic factors (temperature, accumulated rainfall, and relative humidity) in triggering and/or aggravating asthma remains unclear.
In winter, cold air can affect lung function in asthma patients and induce bronchospasm13. Furthermore, rainy spells with less sunshine hours give rise to an increase in indoor humidity, favoring fungal proliferation, and people tend to spend more time indoors in enclosed spaces with others, thereby facilitating the spread of viruses14,15. In addition, greenhouse gas emissions accelerate global warming, causing a rise in winter and spring temperatures and leading to earlier and prolonged pollination seasons. Spores containing allergens interact with pollutants, determining the diffusion and aggregation of gases in the atmosphere7, leading to an increase in human exposure to these gases and thereby increasing the risk of asthma attack and asthma susceptibility15.
Given that the interaction between urban climate and health is setting specific, understanding local seasonal patterns and environmental triggers of asthma is vital, especially in areas lacking longitudinal studies over long periods.
In view of the above, the objectives of this study were: a) to evaluate the influence of climatic conditions (minimum and maximum temperature, accumulated rainfall, and relative humidity) on hospital admissions for asthma and LRTIs among children and adolescents living in an urban settings between 2002 and 2012 and identify epidemic peaks of admissions for asthma; and b) to compare local seasonal patterns of admissions for asthma and LRTIs.
The findings of this study can contribute to asthma management, prevention, and treatment and improve healthcare delivery and the allocation of resources during epidemic peaks. Furthermore, the inclusion of LRTIs, such as viral bronchiolitis, enables the analysis of the patterns of this disease associated with climate change in children aged 0-4 years, helping to clarify diagnosis in this age group.

MATERIALS AND METHOD
A time series study covering the period January 2002 to December 2012 was conducted in Belo Horizonte, Brazil (latitude 19.9°S and longitude 43.9°W). Belo Horizonte is the capital of the State of Minas Gerais, has a population of 2,375,151 inhabitants, an area of 331.4 km2, and population density of 7,167.02 people/km2 16.
The climate is subtropical with a wet season (summer) and dry season (winter). The average monthly temperature is 23°C in the summer (December to March) and 19°C in the winter (June to September). Temperature inversions are common in the winter months. Annual rainfall is around 1,450 mm and the prevailing wind direction is east-northeast16.

Admissions for asthma
Data on admissions for asthma was obtained from hospital admission authorization forms provided by the Belo Horizonte City Department of Health. The study included all admissions to public services of children and adolescents aged 0 to 14 years living in Belo Horizonte registered in the database of the Projeto BH-Viva (implemented by the Urban Health Observatory at the Federal University of Minas Gerais) where the primary diagnosis was asthma (International Classification of Diseases – ICD 10, J45 to J46)17. The admissions were categorized into the following age groups: 0-4, 5-9, and 10-14 years.

Lower respiratory tract infections
Data on admissions to public services for bronchiolitis (ICD 10 J21, J21.0, J21.8, and J21.9) in children aged 0 to 4 years living in Belo Horizonte was obtained from the same source mentioned above. The age group was limited to 0 to 4 years because it is known that during the study period 98% of registered admissions for bronchiolitis occurred in children in this group.

Asthma and bronchiolitis hospitalization rate
To calculate the annual hospitalization rate, the population figures were adjusted to take account of omissions in the 2010 Census conducted by the Brazilian Institute of Geography and Statistics (IBGE)16. A correction factor of 14.6% was used based on omission rates of 15 and 16%, respectively, for boys and girls under five years of age, and 13% for the five to nine years age group.
In addition, we calculated the annual population growth rate based on demographic changes in delimited areas such as the “formal city” and slums. The cohort growth rate during the period between censuses was estimated based on the comparison between the base population in 2010 and population in 200017. The rate of admissions for asthma and bronchiolitis were calculated for each age group based on the number of admissions per month and per year.

Climate data
Data on mean monthly minimum and maximum temperature (°C), relative humidity (%), and accumulated rainfall (mm) was obtained from the Conventional Station of Belo Horizonte (World Meteorological Organization - WMO: 83587), belonging to the National Institute of Meteorology, was provided by the Climatology Laboratory at the Pontifical Catholic University of Minas Gerais and Belo Horizonte City Council18.

STATISTICAL ANALYSIS
Descriptive statistics
The annual rates and monthly admissions for asthma and bronchiolitis were stratified by age group and analyzed against the climate data.

Correlation and regression analysis
The relationship between the rates of admissions for asthma and bronchiolitis and the climate variables was determined using Pearson’s correlation coefficient. The effects of each climate variable were analyzed using Poisson regression, where the dependent variables were rate of admissions for asthma among the 0-14 and 0-4 years age groups and rate of admissions for bronchiolitis among children aged between 0-4 years. To determine whether there was a delayed response of hospitalization rates to climatic conditions, the climate variables were lagged by one and two months before the date of hospitalization.

Time series
An Auto-Regressive Integrated Moving Average (ARIMA) model19 was used to capture seasonal patterns in the variables over time from past values. We used the seasonal model ARIMA (p,d,q) x (P,D,Q)[S], where p is the order of the autoregression (AR), q is the order of the moving average part (MA), d is the number of times the series is differenced, P is the order of the seasonal-AR process, Q is the order of the seasonal-MR process, and D is the degree of seasonal difference19. Trends in the number of admissions in the years 2013, 2014, and 2015 were analyzed together with their confidence intervals. The differences between the orders d and D are applied to remove trends and seasonal effects to make the time series data stationary.
The model was validated by forecasting the number of admissions for asthma for 2011 and 2012 using the ARIMA model without these trends and seasonal effects. Predictive capacity was measured using mean absolute error (MAE), a common measure of forecast error calculated using the following equation:

(Equation)

where is the number of observed admissions, is the number of admissions forecasted by the adjusted ARIMA model, and n is the number of forecasted observations, which in this specific case corresponds to 24 observations.
Data processing and analysis was performed using the statistical software packages R and STATA adopting a significance level of 0.05.
The study was approved by the Research Ethics Committee at the Federal University of Minas Gerais (CAAE 11548913.3.0000.5149).

RESULTS
Rate of admissions for asthma and bronchiolitis
There were 32,978 admissions of children and adolescents (0-14 years) for asthma during the study period, 18,962 (57.5%) of which were males and 14,016 (42.5%) female. Hospitalization rates according to age group are shown in Table 1.
Table 1 shows that the rate was highest in the 0-4 years group (25,926), followed by the 5-9 (5,817) and 10-14 (1,235) years age groups. The rate of admissions for asthma fell during the study period across all age groups, whereas the rate of admissions for bronchiolitis increased.
The peak months for admissions for asthma and admissions for bronchiolitis were March, April, and May, and May, June, and July, respectively (Table 2). The Pearson’s correlation coefficient showed a positive significant association between the numbers of admissions for asthma and the numbers of admissions for bronchiolitis in the 0-4 years age group (0.30, p=0.001).
Table 2 shows that there are two well-defined seasons: a rainy season, characterized by an increase in temperature and humidity; and a dry season, characterized by lower relative humidity. Temperatures vary slightly between seasons.
As expected, the Pearson’s correlation coefficients show a significant correlation between maximum and minimum temperature (r=0.83, p<0.01), relative humidity and accumulated rainfall (r=0.68, p<0.01), accumulated rainfall and minimum temperature (r=0.55, p<0.01), relative humidity and minimum temperature (r=0.45, p<0.01), and accumulated rainfall and maximum temperature (r=0.28, p<0.01), and a very weak negative nonsignificant association between maximum temperature and relative humidity (-0.03, p=0.76).
The Poisson regression analysis of the association between rates of admissions for asthma and climate variables showed that there was a 1% decrease in the monthly admissions rate for every 1mm increase in rainfall and a 5% rise in the admissions rate for every 1% increase in relative humidity in both age groups (Table 3). These associations were significant (p< 0.05).
With regard to bronchiolitis, the results showed that there was a 1% decrease in the monthly admissions rate for every 1mm increase in rainfall and a 21% decrease in the monthly admissions rate for every 1ºC increase in maximum temperature (p<0.05) (Table 3).
To determine whether there was a delayed response of rate of admissions for asthma to climatic conditions, we analyzed the correlation between rate of admissions for asthma and climate variables lagged by one and two months. The results show a significant correlation between rate of admissions for asthma and relative humidity and higher minimum temperature both one and two months before hospitalization (r=0.41 and 0.50, respectively, p<0.01 and r=0.36 and 0.48, respectively, p<0.00), suggesting a possible delayed response.
Poisson regression was used to test the effect of exposure to variations in climatic conditions one and two months before hospital admissions. The results showed an association between admission rates and mean minimum and maximum temperatures, relative humidity, and accumulated rainfall both one and two months before admission. Given that the values for minimum temperature depend on the values for maximum temperature, the analysis was performed adding the variable temperature range, which refers to the difference between monthly maximum and minimum temperatures.
The results of the final model with climate variables lagged by one month showed that there was a 1% decrease in the rate of admissions for asthma for every 1mm increase in rainfall and a 4% rise in rates of admissions for each 1% increase in relative humidity for both the 0-14 and 0-4 years age groups. With regard to bronchiolitis, there was a 1% decrease in the rate of admissions for every 1mm increase in rainfall and a 24% decrease in rates of admission for each 1ºC increase in temperature range (Table 4). These associations were statistically significant.
In the model with variables lagged by two months, only increased relative humidity showed a statistically significant association with rate of admissions for asthma (p<0.05), with a 3% rise in rates for each 1% increase in relative humidity. With regard to bronchiolitis, there was a 19% reduction in rates for each 1% increase in temperature range, suggesting that a reduction in temperature leads to an increased risk of hospitalization (Table 4). These associations were statistically significant (p<0.05).
Time series modeling of admissions for asthma (0-14 years) showed that the best model was ARIMA (0,1,2) (2,0,2)[12], which considers a 12-month seasonal pattern.
To validate the model, the number of admissions for asthma in 2011 and 2012 were excluded and the forecasts for these years were performed using the seasonal model ARIMA (0,1,2)(2,0,2)[12]. The MAE was equal to 55.67 admissions, suggesting that the model adequately predicted the occurrence of asthma within the 95% confidence interval, as can be seen in Figure 1.
The model confirms that there were seasonal patterns in the occurrence of asthma exacerbations over the study period and that downward historical trends point to a reduction in the number of admissions for asthma among children and adolescents living in Belo Horizonte in coming years.

DISCUSSION
Climatic conditions play an important role in various atopic and infectious diseases, which are one of the leading causes of morbidity and mortality in developing countries, particularly affecting children. The main findings of the study can be summarized as follows: (1) the rate of admissions for asthma is higher among children aged 0-4 years and boys and fell over the study period. In contrast, the rate of admissions for LRTIs (based on the rate of admissions for bronchiolitis) increased over the period; (2) the peak months for admissions for asthma and bronchiolitis were March, April, and May, and May, June, and July, respectively; (3) reductions in rainfall and increases in relative humidity are significantly associated with an increase in the rate of admissions for asthma and reductions in maximum temperature and rainfall are associated with rate of admissions for LRTIs; and (4) The model confirms that there were seasonal patterns in the occurrence asthma exacerbations over the study period and that downward historical trends point to a reduction in the number of admissions for asthma among children and adolescents living in Belo Horizonte in coming years.
Asthma is the second leading cause of hospital admissions in children under 14 years in Belo Horizonte City. A study covering the period 1997 to 2000 showed high rates of admissions for asthma together with a downward trend in rates among children under five years20. Another study of the period 2002 to 2012 showed a continuing downward trend, indicating that actions taken to tackle asthma have had a positive impact on admissions6.
However, it is important to stress that despite the fall in admission rates, the number of admissions is high in comparison to developed countries21, 22, suggesting that other factors influence the occurrence of asthma exacerbations.
It is also interesting to note that there was an increase in the number of admissions for bronchiolitis over the study period, corroborating the findings of studies demonstrating that there has been an increase in the circulation of multiple viruses in recent years, particularly affecting children\'s health23. In this respect, the influence of climatic conditions on the multiplication and maintenance of multiple respiratory viruses in urban settings is well-known.
However, the effect of climatic conditions is generally underestimated due to difficulties in assessing individual exposure and lack of information on patient medical history, thereby hindering the measurement of the exposure–response relationship. By understanding the influence of climatic conditions and identifying periods of severe asthma epidemic peaks based on hospital admissions, it is possible to enhance the quality of hospital care and strengthen health promotion and asthma control programs in primary care services.
While a number of studies have shown that there is a relationship between seasonal variations in climate and the prevalence of asthma in Brazil,12,14,24,25 few have investigated hospital admissions for asthma26. Furthermore, these studies present contrasting findings regarding the role climatic conditions play in respiratory exacerbations. Possible explanations for this include variations in climatic conditions and air quality and composition across Brazilian cities and the use of different research methodologies and sampling techniques, thus limiting the generalization of the results presented here.
The present study investigated the association between climatic conditions and cases of severe asthma requiring hospital admission. It is important to highlight there is often a delay between the onset of asthma symptoms and admission to hospital, during which time the patient continues to be affected by the surrounding environment, worsening the condition. Moreover, other factors can contribute to hospital admission, such as poor or delayed access to primary healthcare services and inappropriate or inadequate medication26.
Our findings show that there was an association between increased relative humidity after rainfall and admissions for asthma, with effects persisting for long periods of time. In general, relative humidity refers to the degree of saturation of the air and is strongly influenced by rainfall. However, the present study showed that admissions for asthma were associated with high humidity and low rainfall, suggesting that the delayed effects of rain can lead to an increase in humidity, especially indoors27. Poor ventilation and exposure to the sun can lead to damp and mould and overall poor indoor air quality, contributing to an increase in respiratory diseases. The present study did not investigate the influence of heatstroke on hospital admissions, due to lack of information. Instead, maximum temperature was used as a proxy variable, considering the strong correlation between these variables. However, it is recommended that future studies include this variable.
Belo Horizonte is the most densely populated city in the State of Minas Gerais. Rapid population growth led to unplanned urban sprawl, with the occupation of areas of risk such as steep slopes and stream banks, particularly by low-income groups. Environmental degradation, poverty, and extreme geological risk may contribute to flooding of soils, leading to an increase in indoor humidity. In this respect, a study reported higher rates of admissions for asthma in vulnerable areas such as slums6.
Studies conducted in other states in Brazil confirm the influence of humidity on respiratory diseases in regions with a tropical climate characterized by slight variations in temperature between seasons14, 24,25.
It is important to highlight that, in contrast to the problems experienced in the dry season, during the rainy season high relative humidity combined with the fact that people tend to spend more time indoors increases contact with indoor allergens and spread of viruses8, 27,28. In this respect, exposure to indoor allergens for prolonged periods of time may partially explain why increases in humidity after rainy spells continued to influence hospital admissions for up to two months later.
Only a weak association was found between LRTIs (based on admissions for bronchiolitis) and admissions for asthma in children aged 0-4 years (r=30, p<0.01), suggesting that LRTIs have little influence on severe asthma. In this respect, the timing of epidemic peaks in admissions for asthma and bronchiolitis was different and the occurrence of bronchiolitis was associated with different climatic conditions to those that influence asthma.
The findings show that admissions for bronchiolitis and admissions for asthma followed different patterns. The results of Poisson regression showed a significant association between admissions for bronchiolitis and falls in maximum temperature and lower rainfall, which coincides with winter weather patterns. Studies have shown that breathing in cold air decreases the temperature of the lower airways, facilitating rhinovirus replication29. In addition, in colder weather people tend to spend more time indoors in enclosed spaces with others, facilitating the spread of viruses, especially in schools, crèches, and homes8, 28.
The time series model used by this study accurately predicted future epidemic peaks in admissions for asthma, making it an important tool from a public health perspective. The World Health Organization encourages the development of models capable of predicting disease outbreaks, as they are invaluable tools for tackling and preventing epidemics30.
Despite providing important insights about asthma, this study has limitations, notably that it was impossible to evaluate the role of pollutant concentrations. The main source of air pollution in Belo Horizonte is road traffic, since most factories are located in metropolitan regions and the use of gas heaters is limited. Pollutants may have a toxic effect not only on people with allergies, but also on susceptible individuals31,32. According to the state environment agency (Fundação Estadual do Meio Ambiente - FEAM)33, air pollution concentrations in the city are within legal limits. Nonetheless, future studies should consider the effect of air pollutants.
It is known that rain washes away pollutants, cleaning the air by removing significant amounts of suspended particulate matter. Furthermore, wet soils prevent particle resuspension. On the other hand, wind is slowed by rough surfaces and obstacles, meaning that uneven topography can affect wind speed and hinder the dispersion of pollutants and heat.
In light of the above, future studies should include variables such as wind speed and direction, given that these factors may influence data interpretation, particularly in large cities.
It is also important to bear in mind that the data source was a database that was not designed to inform studies of this nature. Although subject to information bias (including diagnostic bias), the Hospital Information System (HIC) is considered one of the most important tools in the area of public health and is widely used by researchers and decision makers to inform health policymaking, planning and management.
Despite doubts about information reliability, the use of similar databases has increased in recent decades. However, the wide variety of studies that use these sources, coupled with the internal consistency of our results and consistency of our findings with other studies, reinforces the importance of this database and the need to understand its advantages and disadvantages34.
Another limitation is the use of secondary data for admissions for asthma and bronchiolitis without confirming diagnosis and identifying the type of bronchiolitis. Because they have similar clinical characteristics, it is difficult to differentiate between viral diseases and asthma, especially in children under 2 years. However, it is important to stress that viral bronchiolitis requiring hospital admission is an acute condition and its diagnosis is eminently clinical. Patients can develop signs and symptoms that are different to those of asthma and may require a different form of treatment, especially in relation to the response to use of beta-agonists. Patterns of asthma in children under 4 years were similar to those in children aged 10 to 14 years. However, these findings are not sufficient to rule out diagnostic error.
It is important to note that the quality of information provided by health information systems in Brazil, particularly the HIC, has improved significantly in recent years and that the data shows a high level of agreement with the ICD. Studies in this area have shown that the SIH is a valuable tool for epidemiological research and its use, which is still modest considering its potential, should be encouraged. In this sense, the wide-scale dissemination of results could encourage the use of this database34. The dissemination of similar studies is also important so that system managers can become aware of the results and prioritize efforts to improve data quality.
Finally, limiting the study sample to admissions to public health services means that part of the population was excluded from the analysis. However, according to the 2013 National Health Survey, two-thirds of the Brazilian population were admitted to public hospital services, suggesting that the results of this study can be used to make inferences about a large segment of the population.
Although the findings of this study should be interpreted with caution given the methodological approach and limitations mentioned above, they provide valuable insights into patterns of asthma in children and adolescents living in Belo Horizonte and have important implications for health, social, and environmental policy. Interventions focused on asthma prevention, such as health education programs, correct diagnosis of signs and symptoms, and monitoring of severe cases during epidemic peaks, combined with interventions designed to promote immediate access to medication and physical therapy, interventions in and around the home environment, and improvements in socioeconomic conditions, have been shown to reduce the risk of asthma attacks and, consequently, hospital admission1.

CONCLUSION
Climatic conditions play an important role in various respiratory diseases that particularly effect children. Our findings show that the timing of epidemic peaks in admissions for asthma and bronchiolitis was different and that hospital admissions for asthma were associated with a reduction in rainfall and increase in relative humidity, while admissions for bronchiolitis were associated with reductions in maximum temperature and rainfall.
By understanding the influence of climate conditions on the occurrence of asthma and identifying severe asthma epidemic peaks based on hospital admissions, it is possible to enhance the quality of hospital care and strengthen health promotion programs in primary care services. It is necessary to develop effective health promotion, environmental, and urban policies and programs to tackle the underlying causes of asthma and reduce the rate of hospital admissions due to this disease.

REFERENCES
1. Global Initiative for Asthma. Disponível em: http://www.ginasthma.org.
2. Kopel LS, Phipatanakul W, Gaffin JM. Social disadvantage and asthma control in children. Paediatr Respir Rev. 2014; 15(3):256-263. doi:10.1016/j.prrv.2014.04.017.
3. Brasil. Ministério da Saúde. DATASUS - Sistema Nacional de Dados 1993-2010. Disponível em: www.datasus.gov.br.
4. Moura BLA, Cunha RC, Aquino R, Medina MG, Mota ELA, Macinko J., et al. The main causes of hospitalization for primary health care sensitive conditions in Brazil: an analysis by age groups and region. Rev Bras Saúde Matern Infant. 2010; 10(Supl. 1):S83-S91.
5. Bastos RM, Campos SEM, Ribeiro LC, Bastos-Filho MG, Teixeira MTB. Admissions for ambulatory care-sensitive conditions, Minas Gerais, Southeastern Brazil, 2000 and 2010. Rev Saúde Pública. 2010; 48:958-967.
6. Dias CS, Dias MAS, Friche AAL, Almeida MCM, Viana TC, Mingoti SA, et al. Temporal and spatial trends in childhood admissions for asthma in Belo Horizonte, Minas Gerais, Brazil and their association with social vulnerability. Int J Environ Res Pub Health. 2016; 13: 704 doi:10.3390/ijerph13070704.
7. Wang W. Progress in the impact of polluted meteorological conditions on the incidence of asthma. J Thorac Dis. 2016 Jan; 8(1):E57-61. doi: 10.3978/j.issn.2072-1439.2015.12.64.
8. Johnston NW, Johnston SL, Norman GR, Dai J, Sears MR. The september epidemic of asthma hospitalization: school children as disease vectors. J Allergy Clin Immunol. 2006; 117557-62.
9. Silva ECF. Alergia respiratória. Revista do Hospital Universitário Pedro Ernesto, UERJ 2008; 7(2):33-57.
10. O’Connor GT, Neas L, Vaughn B, Kattan M, Mitchell H, Crain EF, et al. Acute respiratory health effects of air pollution on children with asthma in US inner cities. J Allergy Clin Immunol. 2008; 121:1133-9.
11. Grech V, Balzan M, Asciak RP, Anton Buhagiar A. Seasonal variations in hospital admissions for asthma in Malta. J Asthma. 2002; 39(Issue 3):263-26.
12. Silva Júnior JLR, Padilha TF, Rezende JE, Rabelo ECA, Ferreira ACG, Rabahi MF. Efeito da sazonalidade climática na ocorrência de sintomas respiratórios em uma cidade de clima tropical. J Bras Pneumol. 2011; 37(6):759-767.
13. Nielsen KG, Bisgaard H. Lung function response to cold air challenge in asthmatic and healthy children of 2–5 years of age. Am J Respir Crit Care Med. 2000; 161:1805-9.
14. Valença LM, Restivo PCN, Nunes MS. Variação sazonal de atendimento de emergência por asthmaem Gama, Distrito Federal. J Bras Pneumol. 2006; 32(3):284-289.
15. Weiland SK, Husing A, Strachan DP, Rzehak P, Pearce N, ISAAC Phase One Study Group. Climate and prevalence symptons of asthma, allergic rhinitis, and atopic eczema in children. Occup Environ Med. 2004; 61:609-615.
16. Instituto Brasileiro de Geografia e Estatística. IBGE. Censo de 2010. Disponível em: http://censo2010.ibge.gov.br/.
17. Friche AAL, Dias MAS, Reis PBR, Dias CS, Caiaffa WT. Urban requalification interventions and the impact on health: Study protocol “quasi-experimental” with mixed methods–BH-Project Viva. Cad Saúde Pública. 2015; 31(Suppl.):S1-S14.
18. TempoClima PUC Minas. Disponível em: http://www.pucminastempoclima.com.br/.
19. Morettin PB, Toloi CMC. Análise de Séries Temporais. 2ª edição 2006. ABE- Projeto Fisher e Editora Edgard Blucher.
20. Dias MAS, Caiaffa WT, Machado-Coelho GLLM. Poverty is associated with asthma hospitalization and re-hospitalization rates, 1997-2000: An ecological analysis in Belo Horizonte City, Brazil. J Urban Health. 2003; 80: ii108-ii109.
21. Largent J, Nickerson B, Cooper D, Delfino RJ. Paediatric asthma hospital utilization varies by demographic factors and area socio-economic status. Pub Health. 2012; 126:928-936.
22. Brozekm G, Lawson J, Shpakou A, Fedortsiv O, Hryshchuk L, Rennie D, et al. Childhood asthma prevalence and risk factors in three Eastern European countries: the Belarus, Ukraine, Poland Asthma Study (BUPAS): an international prevalence study. BMC Pulm Med. 2016; 14:16:11.
23. Costa LDC, Costa PS, Camargos PAM. Exacerbation of asthma and airway infection:is the vírus the villain? J Pediatr. 2014; 90(6):542-555.
24. Façanha MC, Pinheiro AC. Distribution of acute respiratory diseases in Brazil from 1996 to 2001, Brazil. Rev Saúde Publica. 2004; 38(3): 346-350.
25. Rosa AM, Ignotti E, Botelho C, Castro HA, Hacon SS. Respiratory disease and climatic seasonality in children under 15 years old in a town in the Brazilian Amazon. J Pediatr (Rio J). 2008; 84(6):543-9.
26. Saldanha CT, Silva AMC, Botelho C. Variações climáticas e uso de serviços de saúde em children asmáticas menores de cinco years de idade: um estudo ecológico. J Bras Pneumol. 2005; 31(6):492-8.
27. Jaakkola JJK, Hwang BF, Jaakkola N. Home dampness and molds, parental atopy, and asthma in childhood: a six-year population-based cohort study. Environ Health Perspect. 2005; 113 (3):357-361.
28. Nesti MM, Goldbaum M. Infectious diseases and daycare and preschool education. J Pediatr (Rio J). 2007; 83:299-312.
29. Jartti T, Lee W-M, Pappas T, Evans M, Lemanske Jr, RF, Gern JE. Serial viral infections in infants with recurrent respiratory illnesses. Eur Respir J. 2008; 32:314-320. DOI: 10.1183/09031936.00161907.
30. Global Partnership to Roll Back Malaria. Using climate to predict infectious disease outbreaks: a review. Geneva: World Health Organization; 2004.
31. Tramuto F, Cusimano R, Cerame G, Vultaggio M, Calamusa G, Maida CM, et al. Urban air pollution and emergency room admissions for respiratory symptoms: a case-crossover study in Palermo, Italy. Environ Health. 2011; 10(1):31.
32. Gasana J, Dillikar D, Mendy A, Forno E, Vieira ER. Motor vehicle air pollution and asthma in children: A meta-analysis. Environ Res. 2012; 117:36-45.
33. Fundação Estadual do Meio Ambiente. Relatório técnico: monitoramento da qualidade do ar na região metropolitana de Belo Horizonte no ano base de 2011. FEAM-GESAR-RT-03/2013.
34. Lima CRA, Schramm JMA, Coeli CM, Silva MEM. Revisão das dimensões de qualidade dos dados e métodos aplicados na avaliação dos sistemas de informação em saúde. Cad Saúde Pública. 2009; 25(10):2095-2109.
35. Pesquisa Nacional de Saúde 2013. Disponível em: ftp://ftp.ibge.gov.br/PNS/2013/pns2013.pdf.