Comparison between regression and logistic regression for study population in relation to asthma.
Ambient air is identified by the World Health Organization as a high public health priority, since air pollution associated with increase in mortality and morbidity of disease [1,2]. Respiratory disorders are the most important health problems in Iraq  The reported studies indicated a high prevalence rate of allergic diseases in Iraqi population. In addition, respiratory infection in children represent one of the major infectious problems in Iraq . The community health research that were performed by Tikrit University College of Medicine [TUCOM] have estimated that more than a fourth of Iraqi children with asthma report weekly wheeze and cough. Two-thirds of report school absences and one-third report frequent sleep disturbances due to asthma. 
Respiratory and asthma symptoms are public health challenges in the area of secondary school student’s health and are the leading cause of school absenteeism in children, and result in missed workdays and lost productivity in adults as well . The evidence strongly suggests that poor indoor air quality in schools can impact on the respiratory health of students. Children are at greater risk of the development of respiratory diseases in poor environmental conditions because their immune system is still developing [7-9].
Even though poor indoor air quality (IAQ) may have a role in exacerbation of allergic disorders  the socio-economic status may also have a key role in the development and progress of respiratory symptoms and asthma, especially in school students [11-13]. The influence of socio-economic status could be explained by current and past individual exposures to lifestyle and environmental factors .
Globally allergic diseases form a major health problem with increased incidence with time and thus are with socioeconomic impact on individuals and community . Health care delivery for individuals with allergic diseases is insufficient and/or inadequate. Although, in the last decades there is progress in allergic diseases research, still there is a gap in information related to explore the underlying causes, therapies and eventual prevention [15-17]. Therefore, ‘Global Allergy Forum’. participants concluded that
A new integrative approach is needed to understand how a complex network of immunological, genetic, and environmental factors leads to a complex allergic phenotype .
Clarify the role of environmental and personal factors as risk factors for development of asthma and allergy in secondary school students.
2. Materials & methods
2.1. Site of the study
Kirkuk, an Iraqi governorate that is located North-Western to Baghdad, with 1.200 000 inhabitants. An area that is characterised by production of oil and its products. The governorate is with air pollution, which may be of health hazard in some areas of the governorate .
2.2. Study population
The study was carried out in two sites of Kirkuk Governorate. The first one was a secondary school in Kirkuk city center, while the others (3 schools) located in Kirkuk rural areas. A total of 594 girls included in the study, their age range from 12-22 years, with a mean age of 16.4 years. Of them, 387 (65.2%) were from urban and 207 (34.8%) were from rural areas.
2.3. Assessment of symptoms
The occurrence of symptoms and demographic characteristics was recorded by a self administered questionnaire given directly to each student. The questionnaire requested information on personal factors, health status, physical activities, environmental exposures at present and during childhood, and information on type of residence, type of ownership, house age, size of the dwelling, number of subjects living in the house, type of ventilation, type of cooling, type of heating, presence of animals in the house, presence of tree, grass, spider and wool in houses, presence of cockroach and wall painting.
2.4. Statistical methods
Chi square test was determined using SPSS (version 16) to clarify the significance of differences. Significance of differences in means were calculated by Students t test. In all statistical analyses, two tailed tests and 5% level of significance were used. The influence of different factors on the prevalence of asthma or allergy was analysed by both linear and logistic regression using SPSS statistical package. Odd ratios with 95% confidence interval were calculated from the logistic regression models.
3.1. Asthma for whole study population
For all survey data (combined urban and rural), there was agreement between the two analysis methods in demonstrating significant association between asthma development and risk factors such as oil drinking; oil heating; fan cooling; child respiratory tract infection; child exposure to agricultural dust and work; family history of asthma; child playing; menses irregularity; depression; food allergy; heartburn; IBS; stress; house presence of wool, grass and tree; family history of atopy; and presence of water cycle within house. However, there was agreement between the two methods in relation to negative significant association between crowding index; house ownership; air condition heating; electricity heating and asthma development. Table 1
|Air condition heating||-0.146||0.000||-0.731||0.003|
|Air condition cooling||-0.418||0.067|
|Child exposure to smoking||0.066||0.177||0.348||0.145|
|Child respiratory tract infection||0.186||0.000||0.988||0.000|
|Child exposure to cockroach||-0.022||0.566||0.061||0.766|
|Child exposure agriculture dust||0.171||0.001||1.012||0.001|
|Child agriculture work||0.149||0.002||0.891||0.003|
|Child physical activity||0.055||0.139||0.283||0.145|
|Family history of asthma||0.169||0.003||1.160||0.003|
|Family history allergic rhinitis||-0.033||0.436||-0.187||0.426|
|Family history atopic dermatitis||0.039||0.384||0.342||0.161|
|Family history of atopy||0.150||0.000||0.812||0.001|
3.2. Allergy for whole study population
For allergy development (any allergy) in study population, both models demonstrated agreement of significant positive association with risk factors such as: animal exposure; family history of allergic rhinitis and atopic dermatitis; school stress; child playing; anxiety; depression; psychological problem; irritable personality; house presence of animal, wool, spider and grass; family history of atopy and presence of water cycle within house. However, both models agreed as that allergy development was with negative significant association with BMI; crowding index; house ownership; breast feeding and mother illiteracy. Table 2
|Air condition heating||0.066||0.124|
|Air condition cooling||0.060||0.153|
|Child exposure to smoking||0.414||0.190|
|Child respiratory tract infection||20.361||0.996|
|Child exposure to cockroach||0.123||0.512|
|Child exposure agriculture dust||-0.047||0.370||-0.060||0.823|
|Child agriculture work||-0.059||0.224||-0.215||0.398|
|Child physical activity||-0.052||0.162||0.006||0.975|
|Family history of asthma||0.031||0.571||0.554||0.177|
|Family history allergic rhinitis||0.291||0.000||2.208||0.000|
|Family history atopic dermatitis||0.126||0.004||0.675||0.023|
|Family history of atopy||0.166||0.000||1.023||0.000|
3.3. Asthma for urban study population
When analysis performed for urban population, the two models demonstrated agreement on significant positive association between asthma development and risk factors such as: oil drinking; child respiratory tract infection; child hitting; child exposure to agricultural dust band work; child physical activity; aspirin use; child playing and heartburn. However, agreement between the two models demonstrated a significant negative association between asthma development and risk factors such as: house ownership; air condition heating; electricity heating; air condition cooling; presence of house cockroach, wool, and tree; and family history of atopy. Table 3
|Air condition heating||-0.283||0.000||-1.732||0.000|
|Air condition cooling||-0.258||0.000||-1.946||0.000|
|Child exposure to smoking||0.081||0.179||0.456||0.202|
|Child respiratory tract infection||0.272||0.000||1.176||0.001|
|Child exposure to cockroach||-0.032||0.519||0.111||0.677|
|Child exposure agriculture dust||0.376||0.000||1.855||0.001|
|Child agriculture work||0.365||0.003||3.953||0.000|
|Child physical activity||0.149||0.005||1.338||0.000|
|Family history of asthma||-21.781||0.997|
|Family history allergic rhinitis||-0.124||0.018||0.737||0.027|
|Family history atopic dermatitis||0.102||0.084||0.469||0.230|
|Family history of atopy||0.298||0.000||1.626||0.000|
3.4. Allergy for urban study population
The two models agreement on positive significant association was achieved between allergy development in urban population and risk factors such as: oil heating; animal exposure; child exposure to cockroach; irritable personality; presence of house grass and family history of atopy. However, agreement between the two models was achieved on significant negative association between allergy development and risk factors such as: house ownership; electricity heating; and child residence. Table 4
|Air condition heating||0.085||0.265|
|Air condition cooling||-0.084||0.302||-0.139||0.659|
|Child exposure to smoking||-0.033||0.475||-0.557||0.154|
|Child respiratory tract infection||20.430||0.996|
|Child exposure to cockroach||0.139||0.000||1.0656||0.001|
|Child exposure agriculture dust||-0.055||0.495||22.336||0.997|
|Child agriculture work||0.074||0.442||21.654||0.998|
|Child physical activity||-0.274||0.433|
|Family history of asthma||0.067||0.247||15.993||0.996|
|Family history allergic rhinitis||0.131||0.212|
|Family history atopic dermatitis||0.196||0.288||20.620||0.995|
|Family history of atopy||0.194||0.000||1.506||0.000|
3.5. Asthma for rural study population
In rural population, the two models demonstrated agreement on significant negative association between asthma and risk factors such as: BMI; and crowding index. Table 123. In addition, agreement on significant association was achieved between asthma development and hirsutism. Table 5
|Air condition heating||-0.286||0.082||-38.822||0.999|
|Air condition cooling||0.143||0.273||19.431||0.999|
|Child exposure to smoking||-0.032||0.486||-17.840||0.998|
|Child respiratory tract infection||-41.887||0.998|
|Child exposure to cockroach||-18.718||0.998|
|Child exposure agriculture dust||17.147||0.996|
|Child agriculture work||0.052||0.116||5.250||1.000|
|Child physical activity||12.134||1.000|
|Family history of asthma||-0.094||0.211||-1.864||1.000|
|Family history allergic rhinitis||-0.054||0.394||-23.981||1.000|
|Family history atopic dermatitis||-7.652||1.000|
|Family history of atopy||18.587||0.994|
3.6. Allergy for rural study population
Agreement on significant negative association was achieved between allergy development and BMI. However, agreement between the two models on significant association between allergy development and risk factors such as: family history of asthma; aspirin use; anxiety; presence of house wool and spider. Table 6
|Air condition heating||-0.476||0.106||-0.528||1.000|
|Air condition cooling||-0.095||0.684||-0.944||1.000|
|Child exposure to smoking||20.680||0.998|
|Child respiratory tract infection||-0.110||0.555||-0.592||1.000|
|Child exposure to cockroach||-0.009||0.915||0.065||0.857|
|Child exposure agriculture dust||0.057||0.469||-0.613||0.163|
|Child agriculture work||0.058||0.511||1.488||0.001|
|Child physical activity||0.001||0.984||0.680||0.045|
|Family history of asthma||0.326||0.009||1.535||0.018|
|Family history allergic rhinitis||38.009||0.997|
|Family history atopic dermatitis||-0.158||0.072||-0.066||0.905|
|Family history of atopy||0.070||0.474|
Rates of asthma morbidity and mortality are increasing  and this increase contributed to environmental exposure. Asthma is a complex multifactorial disease in which allergic factors and non-allergic triggers interact, resulting in bronchial obstruction and inflammation . Asthma is the leading chronic disease of children in industrial countries; however, the disease is also common in children in developing countries , and may be extended to involve adolescent. The pathogenesis and underlying causes of childhood asthma is not fully understood, however, early life environmental exposure and life style may be implicated in the etiology of asthma [20,21]. Sensitization induced by allergens is essential step for the development of asthma, however, asthma exacerbation correlated to outdoor and indoor allergens, while indoor allergens influence disease prevalence and severity  However, timing of such environmental exposure during early development may also be important in allergic sensitization and later asthma development . Early exposure to endotoxin from farm environments has been associated with reduced childhood asthma risk , however, endotoxin exposure later in life may increase asthma occurrence especially in agricultural settings .
In the present study, influenza and common cold cause allergic disease exacerbation in 46.8% of secondary school girls. In addition, stress was the predominant (66%) factor that exacerbates allergy in secondary school girls, followed by outdoor air pollution (55.3%), animal exposure (36.2%) and house dust (34%). This finding agreed with literature that implicate viral infections, rather than bacterial infections as exacerbating factor for asthma [26,27]. However, with increasing age asthma exacerbation was mainly associated with other factors such as exercise due to decline in trigger role of respiratory infections with age in children .
Exposure to various constituents including tobacco smoke, wood smoke, air-born allergens, dust mites, mould, and other indoor pollutants is known or suspected to trigger wheezing or exacerbate asthma in children . The level of exposure to these compounds differs in regional Iraq from the situation in developed societies, as children spend more time outdoors with increasing age. Despite the increased exposure to asthma triggers, there are few population-based data examining whether exposure to environmental factors may be associated with asthma in Iraqi adolescent.
Exposure to chemical substances and pesticides exacerbate asthma attack in 53.2% of cases with allergy. Taking these together with air pollution suggest that allergy exacerbated in all cases with these factors, indicated the importance of pollution in the control of allergic diseases. Furthermore, these findings clarify that Kirkuk governorate is an area with high pollution, which warranted application of pollution control program. There is no population based study for adolescent girls in Iraq to compare with. However, there was a population based study in children .
Alsamarai et al  found that exposure to wood, oil smoke, cats, dogs, herbicides or pesticides, and animal and farm environments were associated with an increased risk of asthma among children in Samara city, Iraq. The findings suggest that the aetiology of childhood asthma is complex and may include both early life environmental exposure and early allergic sensitization. Combustion of wood liberates nitrogen dioxide, carbon monoxide, sulfur dioxide and particulate matter, all of which have been associated with increased respiratory illness .
Exposure to oil smoke has been shown to significantly increase the risk of asthma , while particles from wood combustion significantly reduced lung function in elementary school children . The present study indicated that oil heating was a significant risk factor for asthma development in adolescent girls. In contrast, both air condition and electricity heating were with negative impact on asthma development in adolescent girls. The results of this study are consistent with previous observations showing that early transient wheezing and/or increased airway reactivity in children and exposure to products of combustion may be important in the pathophysiology of asthma [20,32,33]. The girls exposure to animal exacerbate allergic diseases. However, animal exposure was not shows a significant association with asthma development for whole data and when sub divided into urban and rural community. Although, animal exposure was an important risk factors for allergy development in urban, rural and whole study population. Alsamarai et al  observed associations between exposure to cats and dogs and childhood asthma which are consistent with other studies [20,34-37], but contrast with other studies which found pets were protective [38,39]. Presence of cats, dogs, sheep and / or cattle with the house were with significant association with asthma development in secondary school girls in Kirkuk. A review of 32 articles suggested anon-significant increase in asthma risk of 11% was associated with the presence of pets in the first two years of life . However, it is difficult to explore the association between exposure to pets and childhood asthma, even in prospective studies, because of issues of temporality and possible confounders associated with keeping pets .
A positive association has been reported between asthma among adults and the use of herbicides and pesticides [41,42], although data on pesticide exposure and childhood asthma are limited . In the present study, exposure to either pesticides or herbicides was associated with an increased risk of asthma in adolescent girls. These results are consistent with report concerning primary school children in Iraq  and other geographical areas [20,43]. Several studies have suggested a reduced risk of asthma with exposure to a farming environment in early life . It has been suggested that exposure to a farming environment causes higher levels of exposure to bacterial endotoxin, eventually leading to the production of several cytokines that shift the balance towards the Th1- over Th2- mediated immunity, thereby reducing asthma risk . In the present study and previously reported one in Iraq , such an inverse association with farm exposure was not evident, as there was a significantly increased risk of asthma in adolescent girls and children with farm-related exposure. In contrast, previous studies have reported that growing up in a farming environment is associated with an increased risk of asthma and that endotoxin exposure may increase asthma risk . The discrepancy between studies may be due to differences in farming practice, crops, lifestyle and other "rural" factors that differ between this Iraqi environment and that in Europe and other regions from which previous reports originated. A further difference in Iraq may be the proximity of stables to the home and time spent in stables ; in this population stables were mostly attached to the family home and sometimes located within the house.
The protective effect of breastfeeding on the development of asthma has raised substantial interest, but the scientific evidence relating to the effect of breastfeeding is controversial . The epidemiological studies have provided controversial results showing negative association consistent with a protective effect, whereas some studies have reported either no association or a positive association between the duration of breast-feeding and the risk of asthma [46-48]. The present study indicated that breastfeeding is with a protective effect on development of asthma and allergy in secondary school girls. In contrast, breastfeeding is a risk factor for asthma development in Iraqi children . Both methodological issues and the complexity of the phenomenon may be responsible for these contrasting results . Differences in several factors, including; the age at which various diseases were experienced, hereditary factors as well as environmental factors may influence the association between breast-feeding and the development of asthma, thus explaining the conflicting results reported to date. The finding of the present study may differ from that reported for developed countries because of variations in the duration of breast-feeding; generally about two years in Iraq. In addition, there is the potential for incorporation of local environmental pollutants into breast milk.
The duration of breast-feeding varies substantially in the reported studies, which becomes critical when fitting the variable if the relation is non-linear as previously suggested . The duration of follow up and the age of onset of asthma are also important, as if breast-feeding could delay the onset of asthma, the prevalence of current asthma would be lower among breast-fed than non-breastfed young children, but similar in later life . There is evidence that hereditary asthma or atopic disease  and exposure to environmental factors can modify the relation between the duration of breast-feeding and the risk of asthma. . The controversial results referred to above may relate to the non-linear relation between the duration of breastfeeding and the risk of asthma .
The finding in this study is of a significant association between food allergy and asthma in adolescent girls is consistent with that reported by others in children . Similarly, the association between a family history of atopy and asthma and developing asthma, with the association higher for asthma than for atopy was consistent with findings of others [4,51,52]. These study findings strengthen earlier reports suggesting that genetics might play an important role in the development of asthma in childhood , with parental asthma being the strongest determinant of asthma. The current study also adds to the literature suggesting that exposure to environmental tobacco smoke increases the risk of adolescent and childhood asthma [4,53].
Reported studies suggest that home environment may act as a risk factor for triggering of asthmatic attach and/or asthma development [54-56], in addition violence may be an asthma attack risk factor . The present study indicated that child hitting by their parents was a significant risk factor for asthma development in urban and rural population when analyzed separately. In addition, stress was a significant risk factor for asthma development in Kirkuk adolescent girls.
Inflammatory mediators released as an outcome of stress and subsequently potentiate allergen induced responses [57,58]. Asthma may be prevented by primary and secondary approaches, however, the physicians mostly relies on performing secondary prevention approach. Our present study indicated that the predominant exacerbating factors are stress, pollution and animal exposure, all can be controlled through a healthcare and social programs and health education.
The present study indicated that smoking was responsible for exacerbation of allergic diseases in 19.1% of adolescent girls. However, child exposure to tobacco smoke is not a significant risk factor for development of asthma and other allergic diseases in adolescent girls. But when the data is collected together, linear regression analyses and not logistic regression analyses, shows a significant association between tobacco smoke exposure during childhood and development of allergic diseases [any one] in adolescent girls. In a previous study reported for Iraq, family history of smoking was associated with asthma (OR=1.52, 95% CI 1.17-1.97; P=0.001) .
The Institute of Medicine concluded that cockroach allergens are causally related to asthma attacks.  Our present study indicated that exposure to cockroach form 12.8% as exacerbating factor of asthma in adolescent girls. In addition, cockroach exposure during childhood was with significant association to development of allergy in Kirkuk population (Linear regression), urban population (Linear and Logistic regression), and rural community (Linear regression). Furthermore, present house presence of cockroach was significant risk factor for development of allergy in Kirkuk population (Linear regression), urban community (Logistic regression), but not for rural community. This could be explained on the basis that the density of cockroach is more in urban than in rural communities. In Kirkuk urban community, present house presence of cockroach was with highly significant association with asthma development (Linear and Logistic regression) in adolescent girls.
Our present study indicated that mold was responsible for 17% as exacerbating factor for asthma in adolescent girls a finding that was consistent to that reported by others [63,66]. By using both Linear regression and Logistic regression models, asthma development in adolescent girls in Kirkuk, Iraq, was with positive association with risk factors that include: oil drinking during childhood, oil heating, fan cooling, child respiratory tract infection, child exposure to agricultural dust and work, family history of asthma, child activity, depression, food allergy, heartburn, IBS, stress, presence of house wool, presence of grass and tree within house, family history of atopy, and presence of water cycle within house. However, when the data of urban and rural communities were analyzed separately, asthma development in urban community was associated with risk factors such as child respiratory tract infections, child hitting by his parents, child exposure to agriculture dust and work, child activity, aspirin use, heart burn, house presence of animal and cockroach, family history of atopy, and ho, family history of atopy, and house presence of wool and tree. The pattern for risk factors for asthma development in adolescent girls rural community was different, indicating that there are differences in risk factors influence between urban and rural population.
Several risk factors have been identified as protective against asthma. The present study indicated an inverse association between crowding index and development of asthma in adolescent girls (Linear regression), urban population (Logistic regression), and rural population (Linear and logistic regression). The same pattern was demonstrated for allergic diseases pooled together.
The present study findings and the reported studies have documented that a decrease in allergic impact of environmental exposure can be achieved by application of specific interventions and subsequently may control asthma attack. However, many children and their families, particularly children who live in poverty and rely on emergency departments as their primary source of health care, and the decline in healthcare delivery in Iraq after the American invasion, may not be receiving adequate counseling about how to avoid environmental exposures. Furthermore, performing a campaigns of educational programs for parents and individual with asthma about environmental controls may play an important role in asthma prevention, control and management [89-91].
To prevent unnecessary exposures to outdoor air pollution, clinicians may provide appropriate guidance to asthmatic subjects and their parents regarding exercise during periods of high pollution. With proper management, many environmental exposures can be decreased. 
Some researchers have shown links between exposure to allergens, pollutants and respiratory symptoms, while in contrast some other researchers have demonstrated that better hygiene and clean indoor environment may contribute to the increased prevalence of allergic diseases and respiratory symptoms. The present study will enhance our understanding and knowledge with regard to the two different hypotheses related to asthma and respiratory symptoms.
The study is significant for several reasons: (1)- Address the influence of different variables on prevalence of respiratory symptoms among secondary school students in Iraq. (2)- Assess the extent to which personal, environmental, socio-economic factors and indoor air pollution will affect the prevalence of respiratory symptoms in school students. (3)- Enhance our knowledge and understanding about the two contrasting theories; the hygiene theory and the theory that higher exposure to air pollutants and allergens is related to asthma and respiratory symptoms. (4)- Summarize the preventive measures to reduce exposure to air pollution and allergens in school environments located in different and also efforts in improving indoor air quality of schools thus reducing the absenteeism and respiratory symptoms in students.(5). Clarify the air pollution impact on health of Kirkuk community.