Diet and Current Asthma Symptoms in School-Aged Children from Oropeza Province -Chuquisaca, Bolivia

Background: Several studies have linked dietary intake and asthma symptoms, nevertheless the results are still inconclusive and data from developing country are scarce. Objective: To assess the association of Food patterns and current asthma symptoms in school-aged children from rural and urban areas of Oropeza Province – Bolivia. Methods: A total of 1655 children attending the fi fth elementary grade were invited to participate in a cross-sectional study in rural and urban areas of Oropeza province. Current Asthma symptoms were defi ned following the International Study of Asthma and Allergies in Childhood (ISSAC) recommendation (written and video questionnaire). Food patterns were defi ned using the ISAAC environmental questionnaire. The Mediterranean diet (MD) score was calculated based on this information. Crude and adjusted logistic regression models were calculated including age, sex and place of living. Results: The age range was between 9 – 15 years (Median 11), and 47.9% were male. Current asthma symptoms were reported in a 17% considering the written questionnaire (WQ) and in a 6% considering the video questionnaire (VQ). Higher consumption of nuts [(p=0.01, aOR= 1.6(1.1-2.2)], corn [(p=0.03, aOR= 1.5(1.0-2.2)] and fast food [(p=0.06, aOR= 1.4(1.0-2.0)] showed positive association with current asthma symptoms. Greater adherence to the MD was associated with lower risk of asthma symptoms. Conclusion: Promotion of healthy food patterns in children is needed in order to avoid the rapid increase of non-communicable diseases in this region. Research Article Diet and Current Asthma Symptoms in School-Aged Children from Oropeza Province -Chuquisaca, Bolivia María Teresa Solis-Soto1,2*, Armando Patiño3, Dennis Nowak4 and Katja Radon4 1Universidad Mayor, Real y Pontifi cia de San Francisco Xavier de Chuquisaca, Sucre, Bolivia 2Center for International Health, Ludwig-MaximiliansUniversity, Munich 3Departmental Service of Health (SEDES) Chuquisaca 4Institute for Occupational, Social and Environmental Medicine, Occupational and Environmental Epidemiology & Net Teaching Unit, University Hospital Munich (LMU), Munich Dates: Received: 09 February, 2017; Accepted: 02 March, 2017; Published: 04 March, 2017 *Corresponding author: Solis-Soto María Teresa, Universidad Mayor, Real y Pontifi cia de San Francisco Xavier de Chuquisaca Sucre, Bolivia, Tel: 591-46461080; E-mail:


Introduction
Asthma is one of the most common chronic diseases in the world [1,2]. Several studies have shown that the prevalence of asthma increases as communities adopt western lifestyles and become urbanized. Since the percentage of urbanization is increasing, it is estimated that there may be an additional 100 million persons with asthma around the world by 2025 [1].
Among the lifestyle factors under discussion to be related with asthma are obesity, physical inactivity, exposure to air pollution and dietary habits. Some characteristics of more urbanized diets which have been related which asthma and atopic diseases includes a decrease in the consumption of saturated fats as butter, but an increase in the consumption of -6 (omega-6) polyunsaturated fat (present in vegetable oils). In the same way, diets patterns with poor consumption of oily fi sh or derived fi sh oil products with high content of -3 (omega-3) polyunsaturated fatty acids might be related with this kind of diseases [3]. Additionally, several studies have found that increased consumption of foods rich in antioxidants (including fruit and vegetables), vitamin C, vitamin E, selenium and carotenoids could have a positive effect in asthma prevention [3].

The International Study of Asthma and Allergies in
Childhood (ISAAC), in consistency with other studies [4][5][6][7], reported lower prevalence of asthma and allergies symptoms in Mediterranean countries [8]. It suggest a potential protective effect of the Mediterranean diet (MD) for asthma and allergies.
In general MD is characterized high consumption of fruit, fi sh, vegetables, legumes, pasta, cereals and grains and at the same time a moderate consumption of milk and low consumption of fast food and meat products [9].
The relationship between diet and asthma is still unclear and inconclusive [4,[10][11][12]. The evidence is even more limited considering developing countries. The present study therefore investigated the association between diet and current asthma symptoms in children from Oropeza Province in Chuquisaca -Bolivia.

Study Design
A cross-sectional study was conducted in rural and urban areas in the north western part of Chuquisaca, Bolivia (Oropeza Province) from July to December 2011. ISAAC recommendations to assess asthma symptoms were implemented [13].

Study subjects
Methodology of this study has been reported previously [14,15]. A number of 43 educational establishments were selected randomly and invited to participate from a total of 185 schools listed in the Regional Education Service (SEDUCA -Chuquisaca). Due to access diffi culties schools with less than 20 students were excluded before selection. A total of 36 schools agreed to participate, from them 10 schools where located in rural areas. In a second step, all students in the fi fth grade of elementary school were invited to participate in each of the sampled schools (2584 children). ISAAC project recommendations were considered for sample size estimation.
It considered 1% of signifi cance level and 80% of statistical power [13]. A total of 2340 parents and children (91%) accepted to participate. In this study, complete data analysis was performed including 1655 children (70.7% of the total). A total of 685 questionnaires were excluded because incomplete data in questions exploring food consumption or asthma symptoms.

Study Instruments and variable defi nition
The International Study of Asthma and Allergies in Childhood (ISAAC) proposed an standardized methodology to assess asthma and allergic symptoms in children in order to make the results comparable across countries and over time.
ISAAC developed two approaches to explore asthma symptoms prevalence, it is through the standardized written and video questionnaire (AVQ 3.0) [13], both of them were used in this study.

The validated Spanish version of the ISAAC Written Questionnaire
(WQ) explore lifetime (ever), current asthma symptoms (considering 12 months prevalence) and also severity of asthma, rhinitis and eczema symptoms [16]. For this analysis the question: ''Have you had wheezing or whistling in the chest during the last 12 months?'' was used to defi ne current asthma symptoms. Additionally The ISAAC Video Questionnaire AVQ3.0 (VQ) [17] was implemented following ISAAC recommendations.
The VQ comprised fi ve video sequences of young people suffering asthmatic symptoms in different situations. In this analysis a positive answer to the fi rst scene of the video questionnaire (moderate wheezing at rest) during the last 12 months was used to defi ne current asthma symptoms. Have been reported a high level of agreement between the fi rst scene of the VQ and the question selected from the WQ to assess current wheezing [17][18][19].
The ISAAC environmental questionnaire (EQ) was implemented to explore average frequency of food consumption in the last 12 months [13]. It asks, in 3 point Likert scale (never/occasionally, 1-2 times, and 3/more times), for consumption per week of 16 food groups. Considering this information Mediterranean Diet (MD) score was computed considering consumption of 11 food groups as other studies recommend [5].
Foods considered in favor of MD received a higher score, while higher consumption. It included: fruit, fi sh, vegetables, legumes, cereals, pasta, rice, and potatoes. These foods received a score of 0 for never or occasionally, score of 1 for a frequency reported of 1 or 2 times per week and score of 2 for a frequency reported of 3 times per week.
Foods considered against MD included meat, milk, and fast food. It received lower score, while higher consumption. It means 2 for never or occasionally, score of 1 for a frequency reported of 1 or 2 times per week and score of 0 for a frequency reported of 3 times per week.
Considering these values, MD score is the result of a sum of all foods considered (minimum 0 and maximum 22). It means that while higher the score, greater the adherence to MD.
On the other hand, some variables were taken into consideration as potential confounders: • Exercise: It was explored asking frequency per week of vigorous physical activity.
• Sedentary conduct: It was explored through two questions. Fist one explored hours per day watching television, and the second one asked hours per playing videogames. In both questions a cut-off of 3 hours/day were used.
Body Mass Index (BMI) was calculated taking into consideration World Health Organization instructions [20]. For this we used WHO AnthroPlus software for children (5 to 19 years)and the following categories were used [21]: Normal weight: It included children up to +1 Standard Deviation (SD) according to WHO defi nition. In this group also were included 14 children with ≤ -1 SD (thin).
Overweight: This group included children between > 1 and ≤+2 SDs Obesity: This category included children with > +2 SDs.

Ethical considerations
This study was revised and had the ethical approved of the National Research Ethics Committee working in the University of San Andres in La Paz -Bolivia. Additionally the study protocol was approved by Regional Education Service (SEDUCA-Chuquisaca). Parents or legal guardians of the children received information about the study objectives at least a week before the planned visit to the educational establishment. At the same time the written informed consent form was required. Voluntary participation from children was respected in all steps of research.

Statistical analysis
EpiInfo V. 3.5.3 for Windows was used for double-entry of data. After that congruence checking was performed in order to avoid possible mistakes. For statistical analysis, data was exported to SPSS v.17.
For descriptive analysis, absolute and relative frequencies were reported. In order to assess potential selection bias, children included and excluded in the analysis were compared using chi-squared test (2 test). Median value and range were reported for MD score.
In order to assess the association of food groups consumption per week (1 -2 times/ and 3 or more times versus never/occasionally) and MD current asthma symptoms, crude and adjusted odds ratios and 95% confi dence intervals were computed considering WQ and VQ. The adjusted models considered variables with statistically signifi cance in the bivariate analysis (P value less than 0.05) and reported as confounders in the literature [22]. It included age (three groups: ≤ 10, 11 and ≥12 years), children residence (urban or rural), and sex.

Results
In order to explore potential selection bias, a fi rst comparison was done between children excluded from the analysis (685), and those included. Excluded children were older and reported watch less hours of television (2=0.08) ( Table 1).
Forty eight per cent of children included in the analysis (complete data) were male and reported a median age of 11 considering a range from 9 to 15 years. Current asthma symptoms (last 12 months), showed a prevalence of 17% considering the WQ and 6% considering the VQ. Regarding nutritional status, about 35% of the children were classifi ed with overweight or obese.
Score of the MD reported a median of 13 points, with a range from 3 to 22) ( Table 1). The foods with the highest consumption per week (3 or more times) were cereals, fruits, vegetables, rice, and potatoes (Table 2).
Considering current asthma defi nition by the WQ and VQ, a higher risk for current asthma symptoms was reported for corn and fast food consumption. In the other hand a tendency as protective factor was found for fruit, vegetables, cereal, rice, butter and milk consumption ( Table 2). Inconsistency in the logistic regression models were found for margarine, pasta and nuts consumption. Not all the results reached statistically signifi cance. The results for MD score suggest that a greater the adherence to the MD lower the risk for current asthma symptoms. The results were stronger considering the VQ (Table 3).

Discussion
The present study suggests a protective role of MD for current asthma symptoms (last 12 months) in children attending rural and urban schools of the central part of Bolivia. Positive answer to the question: ''Have you had wheezing or whistling in the chest during the last 12 months?''in the written questionnaire. 5 Positive answer to the fi rst scene of the video questionnaire: Moderate wheezing at rest In the other hand, although the association is weak, high consumption of corn, nuts and fast food seems to increase the risk for asthma symptoms.
Although the ISAAC standardized environmental questionnaire (EQ) has been reported as validated instrument to report current diet considering different cultural contexts [23], it doesn´t explore eating patterns in greater depth including quantity of food, preparation of meals according geographical area, culture and season [24]. In this sense olive oil is a considerable ingredient of MD, unfortunately it is not reported separately in the ISAAC Environmental Questionnaire.
Nevertheless, measurement of olive oil consumption in Bolivia would be diffi cult given the low consumption of this food in the Bolivian diet. Several studies agree on the potential role of exposures to risk or protective factors on asthma during an early stage of life [25]. Unfortunately, it was not possible to explore these aspects in our study, due to the potential memory bias and lack of knowledge about this by children.
It is important to take into consideration that this study included children from different age group than the one recommended by ISAAC [13]. We decided include children attending the 5 th grade of elementary schools (median age 11) because higher courses (children aged [13][14] reported high drop-out percentage especially in public and rural schools [26], affecting in this way the generalizability of our result to the rest of Chuquisaca, and Bolivia. During pilot study we assured that children in this grade were able to understand and complete the questionnaire in a proper way. On the other hand the use of parental questionnaire for children aged 6-7 years as recommended by ISAAC was not considered because a high adult illiteracy especially in rural areas in Bolivia [27]. Some study strengths are related to the representativeness achieved by a high participation (91%). This percentage is quite similar to previous studies using the ISAAC approach in Latin America countries [28].
In order to explore diet, we considered two types of analysis: fi rst, we analyzed individual foods (16 groups) groups and second, we analyzed MD as diet pattern. The second approach, has been reported as a good option in some epidemiological studies [5,29], but also because as being a summary measure (score), allows us to improve statistical power [6].
In the present study, all associations between diet and asthma were analyzed according to a priori hypotheses; therefore tests for multiple comparisons were not used. Has been reported that multiple testing could be problematic in genetic, prognostic, confi rmatory studies, and when the study is of exploratory nature [30].
Previous studies have mainly used the WQ to evaluate the effect of diet on current asthma symptoms. Although the VQ report is not so extended, it has been suggested as very sensitive instrument considering cultural, language or educational differences [18,31]. Is for this reason that we decided include VQ for asthma symptoms defi nition. In Bolivia, considerable sociocultural and educational differences have been reported, especially between urban and rural areas. The large difference found for current asthma symptoms between WQ (17%) and VQ (6%) could be explained by the lack of knowledge, especially in rural areas) of the written terms causing over reporting of symptoms as has been previously suggested [14,32]. But, on the other hand the video scene could explore presence of more severe asthma symptoms in comparison with WQ [19,33]. This argument could explain the differences in our models, although the results do not show consistency in all the cases.
Several studies have reported a positive correlation between high consumption of fast food and asthma symptoms [23].
Although we found a weak correlation, and only considering WQ, it has been hypothesized that this kind of food affects the modulation of immune responses, therefore could trigger asthma [3,34]. Less evidence had been published explaining the role of corn on asthma. It should be explored in more detail, because corn is an essential component on the diet of most Latin American countries and preparation process might present important differences in each area.
Our result showed a tendency as protective role of high consumption of cereal, butter, milk, rice, vegetables and fruit on asthma symptoms. It is consistent with previous studies and explained because their important content of vitamins and antioxidants [22,23].
Correlation between pasta, margarine and nuts with current asthma symptoms were inconsistent might be by chance. Margarine reported the highest percentage of missing values (11%), for that we could assume that children had diffi culties fi lling this question and also it could be possible that they did not recognize properly the difference between butter and margarine. Although, a potential protective role of nuts on asthma symptoms was suggested [35], our results don´t support that. In Bolivia, as in other countries, it is very common that more ingredients (sugar or salt) are added to nuts, so it could be consumed as snack. In this sense some studies have reported salty snacks as potential risk factor for asthma [36].
Evaluation of individual food groups does not take into consideration the combined effects of meals and the potential interaction between them, for that reason some studies are looking for dietary patterns instead of individual foods [3,6].
Considering that MD was assessed by several studies supported by the idea that this diet has a positive effect on host resistance due the high consumption of antioxidants decreasing in that way asthma symptomatology [3,37]. Additionally MD is characterized by high consumption of foods rich in -3 PUFAs and less consumption of foods rich in -6, in this way synthesis of prostaglandin E2 (PGE2) is inhibited reducing susceptibility to allergic diseases [3]. As expected, children in our study reported less adherence to MD compared to other groups in Mediterranean countries as Spain [38]. Unlike these countries, our population reported low consumption of fi sh. It where an important variability in asthma burden among and within countries have been reported, environmental factor, including diet, are an important points under discussion [39].
Although there are few studies reporting asthma symptoms in Bolivia, it also showed a considerable variation. A previous study in Santa Cruz including children 13-14 years, reported a prevalence of Asthma of 13.5% using the written questionnaire [40]. Santa Cruz, located in the west part of Bolivia, is one of the cities with the greatest agricultural production in the country; at the same time is also the city with the highest economic growth and more accelerated modernization. These contrasts are also refl ected in the diet, with a more accentuated combination between typical and imported than in Chuquisaca.
Considering our results, it is important to explore some social, economic and cultural factors that infl uence food patterns. Some studies reported an association between unhealthy food consumption in children and low socioeconomic status [41]. In the same way, higher prevalence of asthma symptoms was found in countries with high rates of parasitic infections, stress and violence, and limited access to asthma treatment [39].
Overall, our study found a weak association between