Risk and Triggering Factors Associated with Bronchial Asthma Among School-Going Children in an Urban City of Coastal Karnataka

Acknowledgements: We thank the school children and their parents/guardians who participated in this study and the block education officer and respective school authorities for permitting us to conduct this study. Our sincere thanks to MS Kotian, who helped us in analysing the data. Funding: Nil Conflict of Interest: None Permission from IRB: Yes 1Dr. Sowmini P Kamath, MBBS, MD, Associate Professor, Department of Paediatrics.ORCID iDhttps://orcid.org/0000-0001-7902-1012, 2Dr. Shrividya Shrishakumar, MBBS. Department of Paediatrics, 3Dr. Animesh Jain, MBBS, MD, Professor and Head, Department of Community Medicine. ORCID iDhttp://orcid.org/0000-00030250-3608, 4Dr. Anand Ramakrishna, MBBS, MD, Professor, Department of Pulmonary Tuberculosis and Chest Diseases (PTCD), 5Dr. Shantharam B Baliga, MBBS, MD, Professor, Department of Paediatrics, Kasturba Medical College, Manipal University, Mangalore, India. All from the Kasturba Medical College, Manipal University, Mangalore, India. Abstract


Introduction
I n India, prevalence of asthma ranges from 0.2% to 31.1%. 1 Asthma, is hyper responsiveness of airways and is immunologically mediated through allergens / respiratory irritants.Assessing risk factors and triggers is necessary to optimise management.Various studies have documented risk factors association with occurrence of asthma in school going children. 2,3,4,5Asthma exacerbations by specifi c trigger factors are a well-known fact and how it affects their quality of life have been studied by Luskin AT et al 6 .Asthma exacerbations by asthma trigger factors have been documented by few studies 7,8,9,10,11 .This study was undertaken to identify the risk and trigger factors of asthma among school children of urban Mangalore.

Material and Methods
All the private schools in Mangalore were listed and two schools (both Kannada and English medium) were selected based on convenience of travel.It was a cross sectional study conducted over two month's duration.Permission was obtained from the Block Education Offi cer and respective school authorities.Informed consent was obtained from parents /guardians.Institutional ethics committee approval was obtained.
All children aged between 6 to 15 years, studying in classes I-X, and present in class on the day of distribution, were given the questionnaire and included in the study.Subjects who refused to participate in the study were excluded.The prevalence of asthma was determined using the standard ISAAC questionnaire . 12.The objective of the study was to determine risk factors and triggers associated with asthma using a structured and pre-designed questionnaire.The sample size was calculated at 1216 assuming the prevalence of asthma at 8%, 2 at 95% confi dence interval (CI), 80% power, and non-response rate of 10%. 13:It is a reliable and valid questionnaire inventory that identifi es the impact of triggers on daily life and perceived control over the triggers both during initial screening and at subsequent follow-ups.As per ATI, the 32 trigger factors have been organized into six subscales viz., exercise, animal allergens, psychological, pollen allergens, infections, and irritants.Psychological subscale included feeling tense, being excited, feeling alone, being angry, depressed mood, arguments, stress at home, unhappiness, intense worries and feeling weak.Feathers from birds, cats and animal hair, were assessed as animal allergens.Pollen allergens were either pollen from trees/fl owers, grass or from weeds.Climbing stairs, running, bicycle, riding, sports, overexertion was included in the exercise subscale.Air pollutants/irritants included exposure to house dust, strong smells, smell of paint, perfumes, sprays, cigarette smoke and exhaust fumes.Having a cold fl u, sinus problems and viruses were assessed in the infectious subscale.Responses to ATI, which describes how often each trigger causes an attack of asthma, in an individual patient were marked on a scale from 1 to 5,exposure ranging from getting 'never exposed' to 'always exposed',(1=Never,2=Rarely,3=So metimes,4=Most of the time,5=Always) 13 .

Asthma Trigger Inventory (ATI)
In addition, we prepared a self-structured questionnaire enumerating the most common triggers relevant to this locality.Based on this additional questionnaire, trigger factors perceived were categorized as pollutants and chemicals, weather and climate, environment and food products categories.
ISAAC asthma Core Questionnaire 12 was used for assessing the asthma prevalence.A self-structured supplementary questionnaire was used to assess the prevalence of risk factors . 14.Perceived triggers of asthma as per Asthma trigger inventory 13 and as per self-structured questionnaire prepared suiting the city environment was collected.Questionnaires in both English and Kannada (local language) versions were used for the study.The versions in Kannada were translated back into English to ensure reliability and validity.The respective parents/guardians with their consent answered the questionnaires at their respective homes.The answered questionnaires were collected the next day from the respective class-teachers.

Current wheezers
Child with symptom of wheezing / whistling in the past 12 months.

Severe asthma
Anybody with >4 attacks of wheeze in the past 12 months, or sleep disturbance for one/ more than one night per week, or speech limitation because of wheeze in the past 12 months, was categorized as having Severe asthma categories.

Risk factor
Any attribute, characteristic or exposure of an individual that increases the likelihood of developing asthma attack.

Trigger factors
Factors/agents that might trigger asthma attack in an asthmatic subject.
Statistical analysis: Analysis was done using SPSS (Statistical Package for Social Sciences) version 16.The results were expressed as means of rates and proportions.The mean, median and standard deviations among the six subscales were calculated.Any signifi cant association was determined by applying the Chi-square test.A p-value <0.05 was taken as statistically signifi cant.

Results
A total of 1011 responses (83.14% response rate) were included.Prevalence of current wheezers was 9 %( 91/1011) with severe symptoms being present in 28.6%(26/91), indicating severe asthma in 2.6% of the study population.Current wheezers were subjected to the supplementary questionnaire to assess risk factors and triggers.
Figure 1 illustrates the frequency of the presence of different triggers as per ATI (has been calculated by excluding 'never' category and taking the rest of the categories to imply presence of the trigger).
Median and mean values of each subscale of ATI were calculated.In the median format, having a cold was the only trigger with a median of 'most of the time' with a standard deviation of 0.876.Table 2 shows the mean values for each of the subscales.The highest mean was for the infections subscale (2.228), and the least for the animal allergen subscale (1.527).
Figure 2 enumerates frequency of common triggers.In the category of pollutants and chemicals, the highest percentage was for fi recrackers at 49.5%.This was followed by factory smoke (29.7%) and fi rewood smoke (14.3%).In the weather and climate category, the highest prevalence was for rainy and winter seasons at 74.7% and 71.4% respectively.It was followed by travel to hilly areas at 35.2%.In the environment category, the highest values were observed for mosquito coil smoke (46.2%), followed by chalk dust (35.2%), cotton bedding (34.1%), dampness/wet surroundings, moulds/fungus in damp places, use of agarbhattis/incense sticks and air conditioners (30% each).The food products category led with fi sh/shellfi sh and seafood at 29.7% followed by groundnuts and peanuts at 23.1%, packed fried items at 22%, Chinese food, chocolates and chocolate drinks at 21%.   17 .Study by Jain A et al 18 , had documented prevalence of asthma to be almost thrice (18.8%) among children with a family history of asthma when it was compared to those with absent family history of asthma (6.3%), the association being signifi cant.Abdallah AM et al 19 20 .In our study, 40.7% of children lived in such houses and no signifi cant association was found with asthma.
Smoking by parents and household members results in passive smoking.Signifi cant association was found between asthma attacks and the exposure to passive smoking by Cheraghi M et al (OR 1.48; p=0.002 4 , by Chhabra SK et al (p<0.01) 17, and by Pokharel PK et.al. (OR 3.33; p=0.004) 20 .Majeed R et al 14 , and Verma R et.al. 21, had reported 38 .5 % and 39% of subjects in their studies respectively to be subjected to passive smoking.Nearly one fourth of children (24.2%) in our study were subjected to passive smoking, but there was no statistically signifi cant association with asthma.High rise commercial and residential buildings in the city, with multiple stories, provide an easy shelter for birds-especially pigeons, to roost and nest.A study by Curtis L et. al. conducted in Chicago determined that considerable quantities of pigeon allergens can be present in wild pigeon infested environments in which pigeons are not kept indoors 22 .The allergenic proteins are found in bird droppings, serum, feathers, skin scales and pigeon bloom.Positive IgE antibody reactions to droppings, feathers and to sera have been found among subjects with Bird Fancier's Asthma 23 .Pigeon allergens may play an important role in worsening asthma in certain urban environments where they harbour, as was observed by Deo S et.al. in a study in Mumbai, which showed increased circulating IgG antibody levels to pigeon dropping antigens and an increase in hypersensitivity reactions to pigeon allergens among the participants 24 .Around 38% of asthmatics were residing in fl ats in our study, however no signifi cant association was found.
Lodge CJ et.al 25 .in his systematic review and meta-analysis study of 89 articles regarding association between breastfeeding and childhood allergic diseases concluded that there is some evidence that breastfeeding is protective for asthma (5-18 years), with weaker evidence for a protective effect for eczema ≤2 years and allergic rhinitis ≤5 years of age, with greater protection for asthma and eczema in low-income countries.Similarly as per 2006 Aboriginal Children's Survey (n=14,170), conducted by Ming Ye et al 26 , exclusive breastfeeding was protective of asthma, when compared with non-breastfeeding (OR 0.59; p<0.0001) children.In our study, exclusive breast-feeding was practiced by mothers in 72.5% children and was not associated with asthma.WHO, in the current guidelines, advocated exclusive breast-feeding for the fi rst six months, but a few scientists believe that introducing potential trigger foods before six months while continuing breast feeding, could protect children against developing allergies (repeated exposure of the immune system at an early age to an allergenic food primes the body to tolerate and does not cause allergy as the child grows older).Anderson HR et.al. 27 in his review article has shown dramatic increase in asthma, eczema in the United Kingdom since 1955, while Venter C et al 28 , documented an increase in the peanut allergy rates.This has coincided with two third reductions in early weaning.Therefore, it was postulated that late introduction of food could promote food allergies.This led to a research question-Is early weaning an effective approach or not, to prevent food allergy in young children?The Enquiring about Tolerance (EAT) study team by Perkin MR et al 29 , conducted a trial with early introduction of six allergenic foods (cow's milk, sesame, cooked egg, peanut, wheat and whitefi sh) at three months of age.It concluded, that early introduction of multiple allergenic foods in normal breast-fed infants prevent food allergy to certain extent and may be dependent on adherence and dose of allergens introduced during the weaning process.Similarly Tromp II et.al. 9 conducted a populationbased prospective cohort study(n=6905 children) in Netherlands and documented wheezing in 31% at age 2 years and in 14% at ages three and four years,while eczema was reported in 38%, 20%, and 18% of children at the ages of 2, 3, and 4 years, respectively.He found that the introduction of peanuts, tree nuts, hen's egg, cow's milk, gluten and soy before the age of 6 months was not signifi cantly associated with eczema or wheezing 9 .Early weaning had occurred in nearly one fourth(24.2%)of our children, and had no association with asthma occurrence.
Pokharel PK et al in his study showed signifi cant association between the pets (dogs and cats) at home and the asthma attacks (OR 5.5;p=0.045) 20.Verma R et. al. 21 reported 35% of asthmatics had pet animals at home.In our study, 18.7% of current wheezers had pets at home with no signifi cant association with asthma.
Trigger factors: The Asthma Trigger Inventory was fi rst developed by Ritz et.al 13 .for adults, but proved to be perfectly reliable and valid for use in the paediatric population by Wood BL et.al 30 .was answered by our study population.'Having a cold' was the most frequent trigger (94.5%), with 47.3% stating it triggers their asthma 'sometimes' and about 2.2% saying it 'always' triggers an attack followed by 'running' at 79.1% (27.5% 'sometimes', 13.2% 'always') and 'house dust' at 69.2% (31.9% 'sometimes', 6.6% 'always').As per Ritz et al 13 , of the six relevant categories of ATI, the most frequently mentioned categories were climate, air pollution/irritants and physical activity.These were reported by more than fi fty percent of patients (n=247) as one major trigger, while at least one psychological trigger was mentioned by approximately one fourth of cases.According to Abdallah AM et.al. 19 exposure to dust (84.6%), playing, physical activity (58.5%), and common cold attacks (56.9%) were the most common triggering factors for asthma exacerbations.Trigger identifi cation has a crucial role in asthma diagnosis and management.Evaluation of triggers exposure would guide the asthmatic patients to avoid implicated triggers for their attacks and thereby improve their quality of life 13 .
The mean and standard deviation values of the trigger subscales of ATI in our study had consistent values similar to a study by Wood BL et al 30  Relevant to this urban city, most common triggers were rainy season (74.7%) and winter season (71.4%), followed by fi recrackers (49.5%) and mosquito-coil smoke (46.5%).Similarly, in a study by Pradeepa PN et.al 31 .asthma exacerbations were higher in the rainy season (June-6.3% and July -5.8%), the climatic conditions being similar to Mangalore city.Dampness, fungus and moulds are rampant in these areas since it is a coastal area with high humidity; in our study 59.4 % of current wheezers lived in such surroundings.As per the study by Cheraghi M et al 4 , the presence of dampness (OR: 1.59, p<0.0001) at home was signifi cantly associated with asthma.Consumption of pasta/noodles, Chinese food and packed fried items (15.4% to 22%) was similar to the rates described in the study by Awasthi et.al.(16.4%)  32 .
Mangalore is endemic to malaria and dengue attacks, thus there is usage of either mosquito coil smoke or repellent in almost all the houses.In our study, 26.4% and 46.2% of current wheezers were triggered with use of repellent liquid and coil smoke respectively.Sharma VP 33 , reported symptoms of asthma in 0.47% (28/5920) of interviewed symptomatic subjects because of using mosquito repellents.Kumar R et. al. 8 monitored mean concentrations of particulate matter during the pre-burning, burning and post burning phases of mosquito coil in the indoor environment and found it to be 259.2μ/m3, 232.4 μ/m3 and 214.0 μ/m3 respectively and concluded there is emission of respirable particulate matter on burning mosquito coil in the indoor environment, which may accumulate overtime and lead onto respiratory illnesses.
Luskin AT et.al. 6 in his study demonstrated in severe asthmatics, exacerbation rates and asthma related quality of life could be improved with avoidance of asthma triggers.Six asthma triggers viz., emotional stress, cold and/or sinus infection, pollens, animals, mould and /or dampness and dust were assessed in his study.As the number of asthma triggers at baseline increased, the total number of exacerbations increased.On univariate analysis, all triggers were signifi cantly predictive for asthma exacerbations, except for pollen and animals.In multivariate analysis, cold and /or sinus infections were most strongly predictive of asthma exacerbations (OR 1.55,p<0 .001)followed by emotional stress (OR 1.42; p<0.001), mould and/or dampness (OR 1.21; p=0.03).Dust, animals, and pollen showed nosignifi cant association 6 .
Peterson MG et.al. 34 found that those patients with exposure to greater number of trigger factors had poorer asthma outcomes.There is improvement in asthma control with reduction in medical needs, if patient exposure to few categories of triggers/risk factors is minimized as stated by current asthma guidelines 35 .Thus, trigger identifi cation is vital in guiding patient and caretaker to reduce the chronic burden caused on their daily life thereby improving the quality of life.Avoidance of asthma triggers needs to be emphasized to patients by the treating physicians.Along with this, treatment as per protocol guidelines to control asthma and prevent exacerbations targeting to improve their quality of life should be the main aim of the physician.School absenteeism, refraining from physical training period and games would decrease.Academic performance and self-esteem in children would improve.
The study had few limitations.Causal association of various risk and trigger factors with asthma could not be ascertained since it was a cross sectional study.The responses to the questionnaire might have been affected by recall bias.Comprehension of questions may not be uniform in the study population.Since asthma is considered a social stigma, this may lead to underreporting.Another limitation of this study is that the risk factors like past history of bronchiolitis, premature delivery, and previous history of mechanical ventilation were not assessed.Physical examination, PEFR, and spirometry were not done as it was beyond the scope of the study due to feasibility constraints.

Conclusion
For childhood asthma, most prevalent risk factors were family history of allergic rhinitis/ asthma, and living in urban areas.Commonly perceived triggers (as per ATI) were highest in infections and exercise group.Most prevalent triggers in the area were exposure to rainy/ winter seasons, fi recrackers and mosquito coil smoke.Identifying risk factors and triggers would aid treating paediatricians in better understanding of disease and individualisation of treatment.

Fig 1 :
Fig 1: Percentage frequency of perceived asthma triggers in the Asthma Trigger Inventory (ATI) in his study on Egyptian children also found a signifi cant association between positive family history of allergy/ allergic diseases and asthma.Thus; it demonstrates genetic infl uence on the disease.Inadequate ventilation at residing places and the occurrence of asthma have signifi cant association as in studies by Mathew AC et al (OR 4.94, in 11-15 years, p=0.001) 3 and by Pokharel PK et al with absence of windows in living rooms (OR 4.03; p=0.041)

Fig 2 :
Fig 2: Percentage (%) prevalence of common asthma triggers as per structured questionnaire suiting the city

Table 1 :
Risk factors present in the current wheezers (n=91)

Table 2 :
Mean and standard deviations among the six subscales of ATI