- Top of page
- Material and methods
Background: During the last decades there has been extensive epidemiological research to explore the increasing prevalence of asthma and allergy in childhood. The worldwide variations in prevalence of these diseases necessitate regional rapports. Furthermore, time-trend analyses with comparable methods are important in order to monitor the rapidly changing prevalence of these diseases.
Methods: Three cross-sectional questionnaire-based studies of asthma and allergy in schoolchildren were conducted in the counties of Troms and Finnmark, in northern Norway in 1985, 1995 and 2000. The two former studies included children from randomly selected primary schools (n = 1794/1985, n = 1432/1995). The latter study was a part of ISAAC-II Europe study (n = 3853). Identical items of asthma and allergy were employed. The analyses comprised only children 9–11 years of age.
Results: The prevalence of asthma was 9.3, 13.2 and 13.8% in 1985, 1995 and 2000, respectively. However, great gender differences were detected; the prevalence of asthma increased in males from 1995 to 2000, from 14.1 to 17.0%, RR = 1.2 (95% CI 1.0–1.5), but decreased in females 1995 to 2000, from 12.3 to 10.5%, RR = 0.9 (95% CI 0.7–1.1). Furthermore, in children with asthma, a changing trend was found in the external factors that perceived symptoms, from typical allergens towards other, unspecific agents. The prevalence of self-reported atopic eczema/dermatitis syndrome (AEDS) was 13.4, 21.1 and 20.8% in 1985, 1995 and 2000, respectively. The prevalence of self-reported allergic rhinoconjunctivitis was in 16.5, 24.7 and 29.6% 1985, 1995 and 2000, respectively, RR (2000/1995) = 1.2 (95% CI 1.1–1.3).
Conclusion: The prevalence of asthma in girls has reached a plateau and even decreased from 1995 to 2000 which is in contrast to the asthma prevalence in boys that tends to continuously increase. The prevalence of AEDS which increased substantially between 1985 and 1995 did not change from 1995 to 2000. However, the prevalence of allergic rhinoconjunctivitis increased steadily from 1985, 1995 to 2000.
Epidemiological research on asthma and allergy in children has during the last decades focused on an increasing prevalence observed worldwide (1). Sooner or later this increase may reach a plateau, and only through repeated comparable studies, time trends in childhood asthma and allergy prevalence can be followed. There has been a significant reduction in the prevalence of reported asthma in Melbourne schoolchildren, whereas the prevalence of eczema and allergic rhinitis has continued to increase (2). Although several 10-year follow-up studies concerning this issue are published, only a few studies have a third comparable prevalence measurement made after another 5–10 years time period. Prevalence of asthma and allergy was studied in schoolchildren in Belmont, Australia: three cross-sectional surveys over 20 years were conducted. The prevalence of hay fever, eczema, atopy, airway hyperresponsiveness or current asthma (defined as recent wheeze plus airway hyperresponsiveness) did not change significantly. These trends were in contrast with the substantial rise in the prevalence of most of these indicators during the period 1982–1992 (3).
We present a time trend study of asthma and allergy prevalence in primary schoolchildren in northern Norway. The present study is based on three separate cross-sectional written questionnaire-based studies orchestrated in 1985, 1995 and 2000. The objectives of this study were to assess whether or not the prevalence of these diseases was continuously increasing during the study period.
- Top of page
- Material and methods
According to this study of 9–11 years old schoolchildren from northern Norway, somewhat new and surprising trends have occurred in the context of prevalence of atopic disorders. In spite that the asthma prevalence in boys tends to continuously increase over the 15 years studied from 1985 to 2000, the prevalence of asthma in girls has reached a plateau and has even decreased from 1995 to 2000. Moreover, the prevalence of AEDS which increased substantially between 1985 and 1995 did not change from 1995 to 2000. However, the prevalence of allergic rhinoconjunctivitis increased continuously among schoolchildren in northern Norway in these cross-sectional studies from 1985, 1995 to 2000.
The increasing prevalence of asthma and AEDS is considered a major worldwide health problem (14, 15). Several epidemiological studies on the prevalence of asthma and allergy in childhood are orchestrated in our region from the beginning of the 1980s to date (7, 12, 16–25). According to the Norwegian lifetime prevalence studies, both AEDS and AR continued to increase during the period 1981–1995 (4, 7, 16). On the contrary, a review of these studies showed that the prevalence of asthma culminated in the beginning of the 1990s after an increase in the 1980s. Furthermore, recent published studies have shown a stabilization of the prevalence rates of allergic diseases among primary school children other than allergic rhinitis (26, 27). The prevalence of asthma and AEDS has even showed declining trends in children from the first half of the nineties (28–30). The rate of hospitalizations for asthma is on the decline or low and stable in children aged 3 years or older in a study from 1984 to 2000 in Rogaland, Norway (31). No increase in hospital admission due to asthma was detected in schoolchildren in Oslo during 1980–1995 (32).
Nevertheless, despite the interesting and encouraging changes to AEDS and asthma prevalence, we did find that the prevalence of AR, which above of all is associated with IgE-mediated allergy, is continuously increasing. Obviously, these differences in prevalence changes between asthma, AR and AEDS in our study are not easy to explain. Atopy and atopic disorders arise from a multifactor origin with an interaction between environment and genetics (33). It is hypothesized that the increased prevalence of asthma and allergy is connected to the wealth increase and a higher prevalence is found in rich families (34). The differences in prevalence rate in 1985, 1995 and 2000 we observe in our study may reflect the changes in standard of living in Norway the last decades; similarly to other industrialized countries, the socioeconomic conditions in Norway improved greatly in the 1960s and 1970s, but reached a plateau in the 1980s. Consequently, as AEDS and asthma occur earlier in life, it is expected that the prevalence of AEDS and asthma will first culminate, and later this will also be the case with AR. On the contrary, the differences between asthma, AEDS and AR in prevalence changes over time may also reflect a better understanding in public in avoidance of offending allergens; Whereas (typical) allergens connected to asthma and AEDS are in part possible to avoid (animal dander/food), the allergens typically causing AR (pollen) are however impossible to avoid. The causative factor when the children perceive asthma symptoms have over the 15 years studied transitioned from traditional allergens to other and unspecific external factors, which may reflect that asthmatics today are a more heterogenic group than ‘yesterdays’ asthmatics. This observation that especially girls with asthma to a lesser extent experience symptoms due to allergens in 2000 than in 1985, may point towards a changing expression of asthma. However, the great difference between girls and boys observed in asthma trends from 1995 to 2000 does not favour an environmental or public opinion (avoidance) explanation to the changes. Yet, physical activity can possibly lay behind the expression ‘other factors’ when asthma symptoms are perceived. We do know that children are increasingly inactive and that asthmatic symptoms are associated with physical activity in children with increased body mass index (35–37).
Although the Bolle/Holt-asthma-questionnaire has been validated and the combination of the two core questions provides a high sensitivity and specificity (0.96 and 0.88, respectively) (7), systematic bias concerning the asthma label must be expected in a cross-sectional, written questionnaire-based design that is employed in these studies. We have no information neither if the children suffer a mild, moderate or severe asthma, nor do we know about the use of antiasthmatic medication. Furthermore, in a cross-sectional setting, it is impossible to do judgement about causality. This is especially true in case of the finding of the trend changes in asthma symptoms due to allergens and ‘other factors’. We do not know if these changes are due to increasing knowledge about allergy and allergens in families with allergies (and hence avoidance) or if the population of asthmatic children over the 15 years studied has transitioned in direction of a greater percentage of children with mild, ‘intrinsic’ asthma.
The prevalence of asthma is difficult to follow over time owing to changes in diagnostic practice and information bias may explain the trends (38). There is a methodological difference between the two former and the latest (ISAAC) study that is a potential source of bias. The ISAAC questionnaire was more profound with multiple questions and the core questions employed in this study did not have the same central position like in the two former studies. This may be an explanation to the differences in missing answers to core questions. Additionally, the response rate has been declining during our study period. An interpretation may be decreasing motivation in the public to participate in such epidemiological studies. A lack of motivation may also be partly responsible to the high rate of missing data to the asthma questions in the 2000 (ISAAC) study. However, some of the missing data can be explained by uncertainties by the responders, an expression of doubt whether the child is asthmatic or not. Lastly, the trend of falling response rate from 1985 to 2000 must inevitable decrease the reliability of the latter studies.
The high rate of missing responses to the asthma core questions in the 2000 (ISAAC) study could potentially be handled by three ways; by excluding the missing data from analysis, by employing an interpretation that the missing data represents no disease and hence can be treated as such, and lastly, by applying the additional asthma-question available in the ISAAC questionnaire and subsequently reduce the number of missing. In our opinion, excluding all subjects with missing responses to the asthma core question would result in serious misinterpretation of the asthma prevalence in the 2000 (ISAAC) study and lead to a false high prevalence. Epidemiological studies on prevalence of disease in persons who do not complete questionnaires have shown that this mode of response implies no disease (Professor Eiliv Lund, personal communication). In cases of missing answers to core questions we therefore could had chosen to interpret these values as nonaffirmative in order to avoid false over reporting. However, we choose the third and applied the additional asthma-question available in the ISAAC questionnaire and thereby reduced the high rate of missing without obvious misinterpretation of the data and the results.
We conclude that the prevalence of childhood asthma shows diverging trends in northern Norway; although it tends to increase in boys, the prevalence has ceased in girls from 1995 to 2000. Furthermore, our study indicates that after a substantial increase in the prevalence of AEDS from 1985 to 1995, no further increase has occurred from 1995 to 2000. However, the prevalence of allergic rhinoconjunctivitis has been continuously increasing over the 15 years studied.