Prevalence and factors associated with asthma in students from Montes Claros, Minas Gerais, Brazil

1 Departamento de Fisiopatologia, Universidade Estadual de Montes Claros (Unimontes). Campus Universitário Professor Darcy Ribeiro. Av. Rui Braga s/n, Vila Mauriceia. 39401-089. Montes Claros MG Brasil. magnaqcoelho@gmail.com 2 Mestrado Profissional em Cuidado Primário em Saúde, Unimontes. Montes Claros MG Brasil. 3 Departamento de Pediatria, Escola Paulista de Medicina, Universidade Federal de São Paulo. São Paulo SP Brasil. Prevalência e fatores associados à asma em escolares de montes Claros, mG, Brasil


Introduction
Allergic diseases are defined as global health problems affecting both developed and non-developed countries, and are responsible for individual suffering and high socioeconomic burden 1 .In particular, asthma is the most common chronic disease of the respiratory tract in children 2 , causing morbidity and decreased quality of life for both patients and relatives 1 .There have been many recent advances in the understanding of asthma etiology and the mechanisms involved in asthma pathophysiology, as well as the development of new treatments.However, studies conducted in different countries suggest that asthma prevalence is increasing in children and teenagers 3 .
Asthma is a serious public health problem, with both social and economic impacts, since it overloads health services.Recent estimates indicate that asthma affects 334 million people worldwide 4 .In Brazil, asthma is diagnosed in over 20% of children and young people between 7 and 14 years 5 .The age of asthma development can be determined by genetic factors in 34% of cases and by environmental factors in 66% 6 .However, as the innate and adaptive immune responses remain immature in pre and postnatal life, individuals are more susceptible to asthma and similar diseases during childhood 7 .
Epidemiological studies are often used to monitor prevalence trends, to determine the associated factors and severity of allergic diseases, especially asthma and allergic rhinitis [8][9][10] .Classic risk factors for asthma are a family history of allergic diseases, sensitization to environmental allergens (including aeroallergens), exposure to endotoxins, fungal and viral respiratory infections in early life.However, these factors do not fully explain the onset and chronicity of asthma 11 .Other factors that have also been associated with asthma prevalence include male gender, low birth weight, maternal or household smoking and secondary exposure to pollution in urban areas 11 .Moreover, asthma can also be aggravated by climate changes, exposure to irritant chemicals, physical exercise and emotional factors 7 .As a means of verifying and recording trends in the prevalence and severity of asthma and allergic diseases in different countries, ISAAC (International Study of Asthma and Allergies in Childhood) has developed specific research instruments 12 .The use of a standardized ISAAC questionnaire minimizes variations in asthma prevalence estimates, allowing comparison across different locations 12 .In Brazil, this instrument was translated into Portuguese and both reproducibility and validity have been verified 13 .
Despite advances in data recording aiding the estimates of asthma prevalence in Brazil, regional data are still scarce.To date there has been limited capacity to identify the magnitude of disease burden within specific populations and compare the results with previously obtained data from other regions of the country 9 .Within this context, the present study utilized ISAAC questionnaires to investigate asthma prevalence and factors related to its development in schoolchildren living in central and peripheral areas of the city of Montes Claros, Minas Gerais.

methods
A cross-sectional study was performed, carried out in two sequential and dependent stages.Schoolchildren aged from 6 to 14 years old, living in central and peripheral areas of Montes Claros, Minas Gerais, Brazil and who were registered with the Family Health Strategy (FHS) program were eligible to participate in the first stage of the study.
Initially, a standard written questionnaire (WQ) developed by ISAAC was administered in a version validated for use in the Brazilian population 10 .Between the end of 2007 and the beginning of 2008, 1,240 WQs were distributed, with 1,131 of them being considered valid, as they were answered properly (8.8% loss).These questionnaires were forwarded to parents and/or guardians responsible for the students and were administered during FHS medical meetings or by the research team directly at the recipients' homes.Questionnaires were collected after a maximum of five days.Written informed consent on behalf of the student was considered an inclusion criterion.
The WQ contains a field requesting a brief description of student personal data and income, in addition to three sections specific to asthma, allergic rhinitis and eczema.With respect to asthma, the evaluated topics were: cumulative prevalence, active asthma condition, medical diagnosis, associated symptoms, severity, and association with other allergic diseases (rhinitis, eczema and rhinitis, or eczema alone ).The prevalence of accumulated asthma was evaluated by the percentage of affirmative answers to question 1 of the asthma module: "wheezes at some point in your life?".The prevalence of active asthma was estimated by the percentage of affirmative responses to question 2: "wheezes in the last 12 months?".The prevalence of asthma as diagnosed by a physician was assessed indirectly by affirmative answers to question 6: "asthma any time in life?".When necessary, the technical terms and any other doubt regarding the questions were explained during the interview.
In accordance with the answers to the WQ, a second stage case control study was performed, where students were divided into asthmatic (A) and non-asthmatic (NA) groups.Students providing positive answers to question 2 of the ISAAC questionnaire, which asked about "the occurrence of wheezes in the last 12 months" were considered asthmatics.Potential factors associated with the occurrence of asthma in the population were investigated, using the complementary questionnaire (CQ) from ISAAC phase II 10 in both groups.The CQ is comprised of 33 questions related to personal data, environmental, dietary and family conditions, diet, infections and immunizations.
Sampling during the second stage was performed with the purpose of keeping the NA/A ratio close to two to one.A 20% in controls prevalence for each factor associated with asthma , 1.5 odds ratio (OR), alpha error of 5% and test power of 85% 5 were assumed.Therefore, all 230 schoolchildren classified as A were included and 460 NA patients were selected by simple random sampling.The final sample, consisting of the selected students whose parents answered the CQ, contained 172 (31.2%) individuals within the A group and 379 (68.8%) within of the NA group.Similarities between the A and NA groups, in terms of distribution of variables such as gender, residential location (peripheral or central area) and maternal education, was confirmed using chi-square tests.P-values < 0.05 were considered significant.
The study also included a skin test for immediate hypersensitivity (STIH) and a parasitological stool test (PST).The STIH included a standard battery of aeroallergens (FDA Allergenic®, Brazil): Dermatophagoides pteronyssinus (Dp), Blomia tropicalis (Bt), German cockroach (Gc), American cockroach (Ap), dog epithelium (De), cat epithelium (Ce), pollen mix (Po), fungi mix (Fa), positive control (histamine, 10 mg/mL) and negative control (excipient, salt solution).The test was performed on the anterior surface of the forearm and allergens that induced papule formation with an average diameter equal to or greater than 3 mm were considered positive 14 .
The initial parasitological test was carried out by processing the students stool samples using the Hoffmann spontaneous sedimentation technique 15 .The students who tested positive for larvae or eggs of the following helminths were considered to have a parasitic infection: Strongyloides stercoralis, Ascaris lumbricoides, Ancylostoma duodenale, Necator americanus, Taenia solium and Schistosoma mansoni.
Obtained data were entered into an Excel® spreadsheet and statistical analysis was conducted with SPSS® software (Inc, Chicago, IL).The relative and absolute frequencies of studied variables were determined.The association between asthma prevalence and independent variables (personal characteristics, environmental characteristics, eating habits, and health conditions) were tested by bivariate analysis and multivariate analysis using non-conditional binary logistic regression, with estimated OR and respective 95% confidence intervals (CI 95%) reported.Only variables having p-values < 0.20 in bivariate analysis were included in multivariate analysis, and were assessed using the stepwise backward selection procedure.The final model was comprised of factors associated with asthma occurrence (at 5% level) and/or those kept as control variables.The Hosmer-Lemeshow test was used to determine the quality of the model fit to the observed data 16 .
The study was approved by the Ethics Committees of both São Paulo Federal University and Montes Claros State University.All parents and/ or guardians provided written informed consent.

results
The prevalence of asthma, rhinitis and eczema symptoms within the group of 1,131 students answering the WQ during the first stage of the study were assessed according to residential location (center or periphery) and are shown in Table 1.The prevalence of asthma based on symptoms and medical diagnosis, as well as the prevalence of flexural eczema was higher among students from peripheral areas, while cases of rhinitis mainly occurred in individuals from the central area (Table 1).
Variables with p-values below 0.20 were chosen for integration into the multivariate model.

Discussion
The development of asthma is multifactorial in nature and is associated with both genetic and phenotypic heterogeneity, making it challenging to compare the results from different asthma studies 17 .Asthma presents as a complex interaction between environmental stimuli and immunological responses 17 .In this study, the application of an ISAAC standard questionnaire aided the identification of factors associated with asthma prevalence, including attendance at kindergarten, household smoking, family history and factors linked to the immunological system, such as rhinitis and positive responses to STIH.Although the data were collected in 2007-2008, it should be taken into account that the prevalence of asthma in Brazil has been a significant health problem for many years 5,18,19 .The results from this study are therefore likely to have current relevance.In addition, this work addresses a gap in knowledge regarding asthma epidemiol-ogy in Montes Claros, and in future will serve as a reference for the adoption of control measures for asthma in this region.
Epidemiological data shows that children born and raised in rural environments are exposed to a wider diversity of microorganisms such as fungi and bacteria.This results in the development of stronger innate and adaptive immunity, including increased activity of regulatory T cells, which act as mediators of such protection 20,21 .In this study however, children from peripheral areas, who may be comparable to children from rural areas, presented with higher incidence of asthma and eczema.Usually, socioeconomic conditions in these areas, such as poor sanitation, substandard living conditions and large reservoirs of inhalant allergens contribute to disease development 22 .Despite the urbanization of rural and remote areas in Brazil and other places in Latin America, with the aid of social and health programs and broadening of infrastructure services 23,24 , there are still discrepancies in Kindergarten attendance was a personal characteristic assessed as a possible factor associated with the incidence of asthma in the investigated population.Other studies report that kindergarten attendance may contribute to development of the immune system and, therefore, can act as a protective factor against the development of childhood asthma.It is likely that kindergarten attendance promotes interactions between children of different ages in an environment with a high diversity of microorganisms 25 .In this study however, the chance of asthma development in children who attended kindergarten was 1.67 times higher compared with those who did not.This suggests that attending kindergarten is not a factor able to solely explain the emergence of childhood asthma.Further investigations will be necessary to ascertain what conditions are associated with the development of asthma in children attending kindergarten.
Among the environmental factors studied, the only factor associated with higher prevalence of asthma was the presence of household smokers, which increased disease prevalence by 1.53fold.Tobacco smoke is an environmental pollutant of high impact since it is composed of several toxic, carcinogenic and mutagenic substances, which negatively affect the health of both active and passive smokers 26 .There are controversies regarding the association of age with vulnerability to tobacco smoke, however some studies indicate that maternal or familial smoking causes the development and/or exacerbation of asthma 10,27 .In the pre and postnatal phases, it is known that tobacco smoke leads to the development of asthma through induction of interleukin (IL)-13, which stimulates immunoglobulin E (IgE) production, infiltration of inflammatory cells and bronchial hyper-reactivity 27 .This knowledge reinforces the necessity of compliance with the World Health Organization recommendations to utilize legislation and public education to ensure completely tobacco free environments.Such actions would assist in reducing the negative impact of smoking on children's and young people's health 26 .
The development of asthma is frequently associated with atopic sensitization, which can also contribute to rhinitis and eczema development 28 .In this study, asthma was positively associated with the occurrence of allergic rhinitis, as previously reported in other epidemiological studies 29,30 .In our work, prevalence of allergic rhinitis increased the chance of asthma development by 3.35-fold, which is consistent with previous literature 31 .Furthermore, the chance of asthma development was 2.48 times higher in children with positive STIH results, which is the most popular method for the diagnosis of allergic diseases, particularly those caused by aeroallergens 17,32 .
Allergies occur most frequently after children reach the age of two years 28 .Among the major sensitizing agents are mites within house dust, fungi, cockroaches and animal epithelial allergens 28 .Such agents, as well as the immunological responses they initiate, are important risk factors for asthma development, and sensitized children have higher chances of developing new or more severe allergic conditions when constantly exposed to specific allergens 33 .
Genetic factors may also be related to asthma, since the prevalence of allergy in relatives, especially first-degree relatives, is considered an important risk factor for asthma development 34 .Confirming this , our results showed that children whose parents, particularly mothers, reported asthma, rhinitis or eczema had a 3.02 times higher chance of developing asthma.However, genetic factors alone cannot explain the prevalence of asthma, and the involvement of geographic and environmental factors should be considered 6,35 .Therefore, the higher association of maternal allergies with asthma development may also involve immunologic factors during pregnancy, genetic polymorphisms, interactions with oxidative stress genes, maternal exposure to environmental factors both inside and outside the home, exposure to tobacco smoke and occurrence of maternal atopy 11,36 .
In conclusion, asthma prevalence in schoolchildren in Montes Claros was not only related to genetic predisposition, but was also associated with individual history, social demographics, exposure to pollutants such as tobacco smoke and positive responses to allergens.These results obtained can be compared to those from other studies, since they were obtained using a standardized ISAAC protocol.A limitation to this study was the cross-sectional design, limiting our ability to identify the time sequences between exposure to risk factors and development of allergic diseases.However, the data are useful for highlighting the role of residential location in increasing the risk of asthma development, and demonstrate a need for planning specific strategies to prevent and control asthma.

table 1 .
Table2shows the results of bivariate analysis assessing relationships between personal, environmental, eating Prevalence of asthma, rhinitis and eczema symptoms in students from Montes Claros, Minas Gerais, according to residential location.
OR: odds ratio; CI95%: 95% confidence interval; * p < 0.05.** Category of reference for the calculation of OR is the "Central" place of residence.

table 2 .
Bivariate analysis of risk factors for asthma based on asthma status.

table 3 .
Factors associated with the prevalence of asthma (multivariate analysis -logistic regression model).