Two cross-sectional studies were performed six years apart (2007 and 2013), in the same 21 randomly selected primary schools in eight districts of Budapest located either in the densely built-up city centre or a leafy suburb. The districts were chosen by simple random sampling by the Central Data Processing and Registration Office of the Hungarian Ministry of Home Affairs. At the initial teacher-parent meetings in September the parents of pupils aged 6–12 years were asked to complete identical ISAAC-based questionnaires, although the studies did not form part of ISAAC.

The numbers of distributed questionnaires in the two years were 6335 and 6441, respectively. Detailed instructions were given by the teachers before completion. The questionnaires were collected immediately after the teacher–parent meetings, or at most a week later. They were completed by the parents of 3933 children in 2007 (response rate: 62.8%); (boys: n=1976; 50.2% and girls: n=1957; 49.8%), and by the parents of 3412 pupils in 2013 (response rate: 52.9%); (boys: n=1637; 48.8% and girls: n=1720; 51.2%).

The study protocol was approved by the Ethics Committee at Heim Pál Children's Hospital Budapest. Informed consent was obtained from all parents. All parents of the patients included in the study received sufficient information.

Details of the questionnaire, including the definitions, have been published previously.6

The core questions from ISAAC Phase I and its methodology were used.2 The questionnaire was translated into Hungarian. It consisted of two main components: a core questionnaire, which included questions about the symptoms of rhinoconjunctivitis, and an environmental one,11 with questions concerning a variety of potential risk factors. Both cross-sectional studies were carried out in the corresponding season, with identical methodologies, thereby allowing a direct comparison.

The prevalence of “diagnosed AR” was determined on the basis of the answers to the question “Has your child had allergic rhinitis diagnosed by a physician?” The parents of those pupils, who answered yes to this (mentioned) question, did not get questions about current AR symptoms. “Current AR” group consists of pupils who had not been diagnosed with AR by physician, but whose parents gave a positive response to the following question: “In the past 12 months, has your child had a problem with sneezing, or a runny, or a blocked nose when he/she did not have a cold or the flu?” In this way there was no overlap between the two groups. The third question related to allergic rhinoconjunctivitis “In the past 12 months, has this nose problem been accompanied by itchy-watery eyes?” A positive response supports the presence of current AR in comorbidity with allergic conjunctivitis.

The prevalence of “cumulative AR” was taken as the sum of the current AR and the diagnosed AR numbers.

The prevalence of diagnosed allergic disease was determined from the responses to the question “Has your child been diagnosed with an allergic disease by a physician?” If the response was positive, the question, “What kind of allergic disease was it?” identified those pupils who suffered from eczema, food allergy or asthma.

The questions considering the risk factors related to the cumulative AR symptoms can be seen in Table 2.

All statistical analyses were performed by a professional statistician using SPSS 15.0 statistical software (SPSS Inc., Chicago, IL, USA). Missing and inconsistent responses were included in the prevalence calculations, but excluded from subsequent bivariate analysis.

The data relating to allergic diseases were characterised with descriptive statistics. The differences between the two time points were explored with T tests, while the Chi-square test and risk-odds ratios (ORs) were used to examine the risk factors for each variable individually with the AR variable. The 95% confidence intervals (95% CI) for the calculated ORs are given in brackets. If a Chi-square test proved significant, the variable was included in the binary logistic regression, in which the AR was the dependent variable. Univariate odds ratio (uOR) computed by crosstabulation examines every single variable versus AR, whereas adjusted odds ratio (aOR) examines those variables in a regression model that significantly affect AR. The results can be seen in the aOR column in Table 2. In Table 2 the cumulative AR n (%) columns show the number and proportion of patients with AR symptoms in the category.

The annual change in prevalence was calculated by taking the difference between the 2007 and 2013 studies’ prevalence values and dividing it by the number of years between the two surveys.

For feather bedding analysis we used T test and ANOVA for the separate four group analysis (Fig. 1). The result was considered statistically significant if p<0.05.

Figure 1.

Feather bedding and family history of atopy. AR+: children with allergic rhinitis symptoms; AR−: children without allergic rhinitis symptoms; family history of atopy+: children with family history of atopy; family history of atopy−: children without family history of atopy; n: number of children.

(0.13MB).

Table 1 demonstrates the trends in the prevalence of current AR, physician-diagnosed AR and cumulative AR in 6–12-year-old children, over the six-year interval from 2007 to 2013. There were significant increases in the prevalence of current AR symptoms, from 14.9% to 23.5% (p<0.001). Prevalence increases exceeding 1% per year (1.43) were recorded. The prevalence increase in the cumulative rhinitis was 1.38% per year (from 26.5% to 34.8%; p<0.001, statistically significant), whereas there was no significant change in the prevalence of physician-diagnosed AR (11.6% and 11.2%).

The prevalences of comorbid current eye symptoms (itchy eyes and lacrimation) were 81.3% and 93.8% respectively in the current AR group in the corresponding time points (Table 2). The prevalence increase in conjunctivitis was 2% per year, which is the highest rate of all the examined symptoms.

The prevalence of physician-diagnosed asthma remained unchanged from 2007 (6.2%) to 2013 (5.7%). There were significant increases (both at p<0.001) in the prevalence of physician-diagnosed eczema, from 10.2% to 15.4%, and in that of food allergy, from 4.8% to 6.2%.

Table 2 presents the findings in 2007 and in 2013 as concerns the relation between the risk factors and the prevalence of cumulative AR.

In the 2013 study, after the multivariate analysis, the significant risk factors for cumulative AR were male gender, a family history of atopy, frequent upper respiratory tract infections, long-lasting disease before the appearance of the allergy and living in a green area.

In the 2007 study, in the multivariate analysis, only long-lasting disease before the appearance of the allergy proved to be such a risk factor.

Both studies with the univariate analysis and the multivariate analysis in 2013 revealed that the odds ratio was smaller for AR prevalence if the pupils were exposed to feather bedding. Feather bedding was significantly lower in children with a family history of atopy (15.1%, p<0.0001) compared with the group where the family history was negative for atopy (20.5%, p<0.0001). Interestingly, the significant difference in feather bedding between atopic or non-atopic families could be observed in both, the subject group with (11.9% versus 16.6%, p<0.000) or without (17.6% versus 21.7%, p<0.003) AR symptoms (Fig. 1).

Our findings are in line with those of previous studies.4,15,16 It seems that male gender may be considered a risk factor for the development of AR in primary school children, although the underlying causes are unknown.

As in several other studies,11,17 a family history of atopy correlated significantly with the prevalence of cumulative AR. Genetic susceptibility enhances the risk of the appearance of allergic diseases.

Frequent upper respiratory tract infections significantly increased the prevalence of cumulative AR. Experimental research results18 support the epidemiological data17 that viral respiratory infection may contribute to allergic sensitisation.

The results of both of our studies revealed that long-lasting disease before the appearance of the allergy significantly increased the risk of the development of cumulative AR. Children have six to eight viral upper respiratory infections per year.19 These infections may be complicated by secondary bacterial infection, and the respiratory tract may be considered a unique morphofunctional entity.20

Both of our studies suggested that the prevalence of cumulative AR was significantly higher among pupils who underwent adenoidectomy, although logistic regression analysis did not indicate this as a significant factor explaining the presence of the AR. Some recent publications21,22 concluded that the presence of AR leads to allergic inflammation around the adenoid tissue, thereby increasing the possibility of adenoid hypertrophy. Statistical analyses demonstrated that the occurrence of adenoid hypertrophy in children with AR peaked in the group of 6-year-olds.21 Our study population (6–12-year-old schoolchildren) may have had problems leading to adenoidectomy before the allergy was diagnosed by a physician.

Both of our studies revealed that the administration of antibiotics given in the first year of life is significantly associated with the development of cumulative AR, as previously demonstrated by the largest study of its type involving the participation of 193,412 children from 71 centres in 29 countries.23 As suggested by the hygiene hypothesis, antibiotic use during infancy can increase the risk of atopic diseases by decreasing the protective effect of the gut microbiotic flora.16

In the studies by Siebers at al., feather pillows were found to harbour lower levels of house dust mite allergens than non-feather pillows, the reason perhaps being that the tighter weave of the pillow cover makes it more impermeable to house-dust mites.7 At first sight, feather bedding may seem to be a risk factor but in our population we found that family history of atopy lessens the habit of feather bedding compared to non-atopic families (15.1% versus 20.5%). In 2007, 23.0%, and in 2013, 26.2% of the children exposed to feather bedding had symptoms of AR as compared with 28.5% in 2007 and 36.2% in 2013 of the pupils not exposed to feather bedding. Our data are in line with other epidemiological studies, but our results add a new question to the epidemiological evidence of bedding and atopic family history: that avoidance of feather bedding could be a result of an atopic family's choice.8–10 It is possible that being aware of the advantages of the avoidance of indoor allergens, the parents of atopic children may have decided not to use feather bedding. Our epidemiological examination measured a smaller feather bedding ratio among patients with a family history of atopy with AR, as well as without AR (Fig. 1).

Visible mould in the bedroom significantly increased the risk of the development of cumulative AR in both of our analyses, supporting the data of a number of epidemiological studies.11,24,25 Home dampness promotes the growth of fungi, their spores and toxins. Home dampness and mould exposure have been associated with the development of allergic diseases.26

In both of our studies, the risk of the development of AR was significantly higher in children who lived in a house with prefabricated concrete walls. Newly built or renovated houses with insulated windows and central heating systems have been reported to have higher levels of indoor air pollutants, chemical products used for cleaning and cockroach allergens.24,27 Our 2013 study demonstrated a positive correlation between living in a green area (leafy suburbs) and cumulative AR. In the green areas of Budapest, there are more grasses, trees and weed species than in other parts. Some studies have suggested that environmental pollution enhances the allergenicity of pollen.28 By attaching to the surface of pollen grains and of plant-derived paucimicronic particles, air pollutants can modify the morphology of these antigen-carrying agents and change their allergenic potential. Consequently, air pollution may not only worsen the symptoms of AR, but may also promote airway sensitisation to airborne allergens in susceptible patients.29

Some limitations should be considered when interpreting our results. The prevalence of AR was measured based on questionnaires, without performing clinical examinations or allergy skin tests to confirm reported symptoms. Current AR symptoms were related to the past 12 months, which may involve further limitations of the study. On the other hand, the reported “itchy-watery eyes” concurrently with the nasal symptoms give support to “current AR”. Thus, the outcomes may have been over- or underestimated. The strength of this study is the comparability of the results. The authors used a standardised method duplicated in a repeat survey in identical schools, identical age groups, surveyed six years apart. In addition, the data set is large enough to reflect the population of Budapest.

The authors declare that no patient data appears in this article.

The authors declare that no patient data appears in this article.

The authors declare that the procedures followed were in accordance with the regulations of the responsible Clinical Research Ethics Committee and in accordance with those of the World Medical Association and the Helsinki Declaration.