SEARCH

SEARCH BY CITATION

Keywords:

  • adulthood;
  • asthma;
  • atopy;
  • childhood;
  • passive smoking;
  • pet

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Few studies investigate how environmental factors in childhood may influence adult respiratory health. The European Community Respiratory Health Survey (ECRHS) is a longitudinal multi-centre study of Western world adults, including retrospective assessment of early life factors.

Analyses of the ECRHS showed both beneficial and harmful long-term effects of childhood factors on adult asthma, allergy and lung function. Childhood pets were associated with less adult atopy and hay fever; beneficial effects were also indicated for growing up on a farm and for early exposure to other children. The findings have contributed to further development of the hygiene hypothesis and further understanding of the mechanisms relating microbial stimulation to allergy; however, the public health consequences may be limited. Harmful effects of early life factors on adult asthma and lung function were indicated for severe respiratory infections early in life, parental smoking and long-term dog keeping. Intervention with regard to parental smoking and vaccination against common lower airways infections may improve respiratory health in the population.

Thus, early life environment had permanent beneficial and adverse effects on adult respiratory health. The multi-centre structure of the ECRHS, the large sample size, the extensive information about each participant and follow-up until the age of 56 years, have given the basis for convincing conclusions, and made possible publication of unsuspected findings in spite of the problems related to adult recall of childhood events. The ECRHS have contributed substantially to increased knowledge about the early life origins of allergy and obstructive pulmonary disease, providing a basis for prevention.

Please cite this paper as: Svanes C. What has the ECRHS told us about the childhood risks of asthma, allergy and lung function? The Clinical Respiratory Journal 2008; 2: 34–44.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Some of the novel concepts in epidemiology during the last years concern the importance of early life environment for health and disease throughout life. Dr Anders Forsdahl at the University of Tromsø in Norway, reported in 1977 and 1978 that arteriosclerotic heart disease in subjects 40–69 years, and cholesterol at age 35–49 years, were both correlated to infant mortality in the municipality of birth (1, 2). David Barker has since developed further and contributed to prove the hypothesis that intrauterine and infant health may determine susceptibility to adult disease (3, 4). Early life origins has also been suggested for obstructive lung disease (5, 6). The Barker hypothesis suggests that environmental influences may be of particular importance during sensitive time widows in organ development. Airways development during intrauterine life, multiplication of the alveoli during the first year of life and lung growth during puberty are believed to be important time windows in the development of obstructive lung disease (7, 8).

Concerning allergy, the importance of childhood environment has been recognised for decades. Research on allergic diseases has mainly focused on sensitisation to allergens and allergen avoidance. The ‘hygiene hypothesis’ suggested by David Strachan in 1989 as explanation for the inverse association of family size with hay fever (9) led to a shift of focus to the role of microbial stimulation in development of allergy. Today, there is quite a lot of evidence suggesting that the natural immune response to commensal and pathogenic microbial agents may promote a non-allergic immunological priming early in life (10–13). Microbial stimulation through gut colonization may be of particular importance (14, 15). Another important development in the field of allergology is the current interest around the role of tolerance induced by natural allergen exposure early in life (16, 17).

There is a rich literature about childhood environment and asthma and allergy in children, but only few studies investigate how environmental factors in childhood influence adult respiratory health. Such knowledge is important both with regard to practical advice to the population about prevention, but also for the understanding of the patophysiology of asthma and allergy. The ECRHS is a longitudinal study of adults from centres from the Western world, including retrospective assessment of early life factors. This review concerns the contribution by 2007 of the ECRHS to the understanding of the early life origins of adult obstructive lung disease.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

The ECRHS cohort was established in the early nineties, including random population samples of young adult men and women aged 20–44 years at inclusion. The study mainly includes centres from Europe, but also centres from other parts of the Western world. At each stage, some centres have withdrawn because of logistic and funding problems. The methods for the ECRHS are described in detail, with protocols and questionnaires at the study website http://www.ecrhs.org; a brief summary is given below (see Fig. 1).

image

Figure 1. The design of the European Community Respiratory Health Study (ECRHS) and the Respiratory Health in Northern Europe (RHINE) study.

Download figure to PowerPoint

Design

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

ECRHS I, stage I: Participating centres were selected from an area defined by pre-existing administrative boundaries, with a population of at least 150 000 people. An up-to-date sampling frame was used to randomly select at least 1500 men and 1500 women aged 20 to 44 years. All subjects were sent a questionnaire enquiring about respiratory symptoms and attacks of asthma in the last 12 months, current use of asthma medication, and nasal allergies, including hay fever. This screening was carried out in 1991–1993, and the sample consisted of 54 centres with 200 682 participants (median response rate 78%)(18).

ECRHS I stage 2: In 38 centres, a random sample of the responders was invited to further investigation, including an interviewer-administered questionnaire and measurements of lung function, bronchial reactivity and serum immunoglobulins E (IgE). Most centres included an additional symptomatic sample in stage 2. Altogether, 21 809 subjects (median response rate 65%) participated.

ECRHS II is the follow-up of the participants in stage 2 of the ECRHS I, taking place from 1998 to 2002. Twenty-nine centres took part in the ECRHS II, which used data collection methods similar to ECRHS I stage 2. Of 14 681 invited subjects, 11 168 (76%) participated. The mean follow-up time was 8.9 years (inter-quartile range 8.3 to 9.5 years).

Ethical approval was obtained for each centre from the appropriate institutional or regional ethics committee, and written consent was obtained from each participant at all occasions.

Assessment of early life factors

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Most information about early life was obtained retrospectively from adults through face-to-face structured interviews in ECRHS I stage II and in ECRHS II (the questionnaires are provided at http://www.ecrhs.org). Information about latitude at birth, language and month/season of birth was obtained from official sources.

Assessment of adult respiratory health

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Questionnaires

Information about respiratory and allergic symptoms, as well as about confounding variables, was obtained through interviewer-led questionnaires (ECRHS I stage II and ECRHS II) or a postal questionnaire (ECRHS I stage I).

Lung function

Forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) were recorded by a standard spirometric method (19). The maximum FEV1 and maximum FVC of up to five technically acceptable blows were determined, and whether FEV1 and FVC each met the American Thoracic Society criterion for reproducibility. The ratio between FEV1 and FVC was calculated as FEV1/FVC ratio. Decline in FEV1 was calculated as ECRHS II value minus ECRHS I value and expressed per year of follow-up. Height and weight was measured before measurement of lung function.

Bronchial reactivity

Methacholine challenge was conducted with a dosimeter (Mefar, Brescia, Italy) (20, 21). FEV1 was recorded 2 min after each inhalation with metacholine, and the test was stopped when either a 20% fall in FEV1 was achieved or a final maximum dose of 2 mg metacholine had been given. The degree of bronchial responsiveness was expressed as ECRHS slope; bronchial hyperresponsiveness was defined as having a 20% fall in FEV1 on methacholine challenge.

Atopy

Specific IgEs were measured using the Pharmacia CAP system (22). During both in ECRHS I stage II and in ECRHS II, blood samples were handled in a similar manner and analysed in a central laboratory (Pharmacia Uppsala in 1992 and King's College London in 2002). Assays for specific IgE were considered positive when in excess of 0.35 kU/L of the specific allergen. Atopy was defined as specific IgE to cat dander, house dust mite (Dermatophagoides pteronyssinus), timothy grass and/or Cladosporium herbarum.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Adult reporting of childhood events

There was hardly any knowledge about adult reporting of childhood events. An analysis of childhood pet keeping in the ECRHS I and II, where participants were interviewed twice on average 9 years apart, showed that long-term reliability in adult reporting of childhood pets was substantial and not influenced by disease status (23). The reliability was higher for a more important childhood event (dogs and cats vs birds), and reliability was influenced by childhood characteristics (i.e. moving house, family size) rather than adult characteristics (i.e. age, social class). For comparison, the reliability was similar to reliability in reporting the building year of one's current home (in decades). Imperfect reliability contributed to underestimation of the effects of pets on adult allergy, i.e. with a kappa of 0.71, a true odds ratio of 0.80 would be attenuated to 0.86. The study concludes that collection of information about childhood pets from adults appears to be reliable for the purpose of studying adult asthma and allergic disease.

Pet keeping

Previous advice to the population concerning pet keeping in childhood – that everybody and in particular those with allergic predisposition should avoid pets in order to avoid allergy – has proved erroneous. The ECRHS I has contributed substantially to this by being the first (24, 25), together with Hesselmär et al.(26), to hypothesise protective effects of childhood pet keeping on asthma and allergy. Svanes et al. observed less adult atopy in subjects who had kept a dog in childhood and suggested that the findings could be explained by microbial stimulation during a sensitive time window in immune maturation (24). Roost et al. demonstrated less adult cat sensitisation in predisposed subjects with a childhood cat compared with increased cat sensitisation in adult cat owners, and suggested that the findings could be explained by the development of tolerance on high-dose exposure to cat allergen early in life (25). This started a paradigm shift after decades of focus on allergen avoidance, and there has since been several studies showing protective effects of childhood pets on various outcomes and in various subgroups, particularly among subjects with allergic predisposition. However, publication bias is likely to have occurred following international excitement, and further analysis of childhood pets and adult asthma and hay fever in the ECHRS I (27) confirmed some protective effects of dog keeping with regard to allergy, and indications of tolerance to high-dose cat exposure, but also showed increased respiratory symptoms related to dog-keeping among the non-allergic and to cat keeping among sensitised subjects living in areas with lower community prevalence of cats (27). The protective effects of childhood pets appeared to be restricted to asthma with onset in childhood (28, 29). One analysis of the ECRHS I in Australia revealed that among subjects with bronchial hyperresponsiveness, having symptoms was more common in subjects who had kept pets in childhood (30).

One main criticism towards the postulated protective effects of childhood pets has been the possibility for selective avoidance, that subjects at risk for asthma or allergy might avoid keeping pets. As good intervention studies of long-term effects of pets are difficult if not impossible to perform, understanding about selection mechanisms is important. An analysis of childhood and adult pet keeping subsequent to asthma in ECRHS I and II (31) showed that some degree of selective avoidance was present for cat keeping in childhood and adulthood, but not for dog or bird keeping. Thus, under certain circumstances, people avoided pets because of asthma and allergy. The clinician should know that this is relatively uncommon; asthma and allergy is often of limited importance for the choice of keeping a pet, and those who already keep pets usually continue to do so. From a scientific point of view, however, the message was that selective avoidance is clearly present and must be taken into account when analysing associations of pets with asthma and allergy. The study concluded that selective avoidance appeared to be of limited magnitude, and most likely accounts for only a part of the described protective effects of pets on asthma and allergy (23).

Farming

The hypothesis that the protective effects of dogs on atopy might be explained by advantageous immunological challenge from dog microbes (24) was developed further through research on farming environment and endotoxins (32, 33). Repeated studies show that subjects who grew up on a farm have less atopy and hay fever, while the literature is not consistent with regard to asthma. There is evidence that some of the protective effects can be attributed to endotoxins stimulating TH1-responses early in life (while the unspecific pro-inflammatory effects of these microbial cell wall products are well known to cause asthma later in life (34). Some centres in the ECRHS I collected data on childhood farm environment, and Leynaert et al. showed protective effects with regard to atopy and hay fever, but not for asthma (35). The particular contributions of this analysis were showing that protective effects of growing up on a farm persisted until adulthood and that these effects were consistent between centres, supporting a biological rather than sociocultural explanation. The study also showed that the protective effects were stronger in subjects of younger generations. Gene-environment interaction analyses from two French ECRHS centres showed that protective effects of farming varied according to CD14C-159T polymorphism, indicating that the ability to respond to microbial stimulation with a positive immune-development varied with genetic status (36).

Family size and day care

The association of family size with hay fever and atopy (37, 38) is remarkably consistent, as demonstrated in the ECRHS I across 36 socioculturally different centres in the Western world (24, 39). The consistency of this finding suggests that it may have a biological explanation, and we believe, according to the hygiene hypothesis, that close contact with other children during immunological priming early in life contributes to an immunological maturation that provides permanent protection against allergy. However, the association of family size with asthma is inconsistent, possibly because of more complex underlying mechanisms reflected in a U-shaped association, as observed in the ECRHS I (39). Typically, an analysis of the ECRHS I in Great Britain showed an inverse association of family size with asthma as well as atopy and hay fever (40), while in some ECRHS centres, the association with asthma was positive (39). Day care constitutes a mixture of exposures, like commensal micro-organisms, infectious agents and allergens. Less allergy in subjects who had attended day care was first shown by Krämer et al., who observed that the effect was only found in subjects without siblings (41). In the ECRHS I day care was associated with less atopy and less hay fever in subjects with no siblings, consistent across centres, but with increased respiratory symptoms, also consistent across centres (39). Roberto de Marco showed that a protective effect of contact with other children (day care or shared bedroom) on diagnosed asthma was found for asthma with onset both during childhood, adolescence and adulthood, and that exposure to children was associated with higher remission rates of asthma as well (28).

Latitude

Analyses of allergic rhinites as related to latitude and birth date in ECRHS I stage I, including 200 682 subjects from 54 centres(18), revealed that the timing of first exposure to seasonal allergens within the first year of life or the timing of infections during winter months during first year of life did not appear to influence adult allergy. Although there appeared to be effects of climate and UV exposure as reflected in significant effects of latitude, language group was found to be the strongest determinant, reflecting so far unknown genetic or cultural risk factors.

Parental smoking

Parental smoking is related to increased risk for asthma and lower lung function in childhood (42, 43), and a few studies address the impact on adult respiratory health (44–50). Analysis of the ECRHS showed that paternal smoking (passive smoking) was associated with respiratory symptoms and lower lung function in adult men, while maternal smoking (passive smoking + intrauterine exposure) appeared to have greater impact on adult respiratory health in women (44). Both findings were consistent across centres with highly variable prevalence of maternal smoking, supporting biological rather than sociocultural explanations. Effects on lung function were seen also in non-wheezers, excluding a major role of differential recall bias. The findings suggest that negative effects of parental smoking on the airways persist until adulthood, and that the time window for particular vulnerability to parental smoking may differ between men and women.

Lower airways infections

The literature indicates that severe lower respiratory tract infections (LRI) in childhood may cause permanent damage to the developing lungs (51, 52). An alternative explanation for the relationship of LRI in childhood and adult obstructive lung disease could be that subjects with poorer immune competence, in particular poor TH1 immunity, may have both increased risk for persistent TH2 immunity and atopic disorders, and poorer defence towards severe infections (53). The literature mostly consists of cross-sectional studies with retrospective information about childhood exposures, but the association is also described in cohort studies of children (54, 55) and in historic prospective cohort studies of adults (5, 6). The issue is addressed in local analyses of the ECRHS I (29, 56). Further, a life-event analysis of the ECRHS I showed how severe respiratory infections before age 5 was associated with more asthma, starting both in childhood, adolescence and adulthood, and with less remission of asthma in all age groups (28). Analysis of hospitalisation for lung disease <2 years in ECRHS I stage I in Bergen revealed stronger associations with previous asthma in men and with current asthma in women, indicating that susceptibility established in childhood may manifest in symptoms at different ages in males and females (57). Both these analyses used methods that to some extent could overcome the problems of recall bias. Analyses of the ECRHS II (58) showed that lower airways infections before age 5 years or hospitalisation for lung disease before age 2 years were associated with lower lung function and more respiratory symptoms at all ages (20–56 years), and with more new-onset asthma in adult life. However, lung function decline was not significantly larger among those with early life airways infections. Subjects with childhood respiratory infections did have higher bronchial responsiveness still in adulthood, but this effect was small compared with effects of adult airways calibre and adult IgE sensitisation(59).

Birth characteristics

There is some evidence that adult asthma and lung function is related to intrauterine environment as reflected in birth characteristics (5, 60, 61, 62). We have information about birth characteristics for parts of the ECRHS: In Bergen, the data from ECRHS I stage I has been matched with the Medical Birth Registry of Norway, and in six Northern European centres (Bergen, Reykjavik, Tartu, Umeå, Uppsala and Aarhus), birth protocols were searched for information about participants in the ECRHS I stage II. In the Bergen study, a linear association between asthma symptoms in young adulthood and birthweight was found within the normal birthweight range also (60). There was an effect of poor intrauterine growth in full-term subjects, and an independent effect of pre-term delivery. However, no association of birthweight with adult lung function and asthma symptoms was found in a large multi-centre study of Northern European (63). No association was found between birthweight and adult atopy (64). Subjects with lower maternal age had more adult asthma, as shown in the Bergen study (60) and in the Nordic Baltic centres (65). In the latter analysis, the association proved to be consistent across centres and consistent when adjusting for a range of potential confounders, suggesting a biological explanation (65).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Summary of results

The ECRHS has contributed substantially to further understanding of the early life origins of adult respiratory health, showing influences of a range of early life factors on adult allergy, asthma and lung function. Analyses of the ECRHS showed convincing and consistent protective effects of childhood dogs on adult atopy(24) and reduced cat sensitisation in predisposed subjects with childhood exposure to cats(25). These studies contributed to a paradigm shift concerning the role of pets in allergy, suggesting that tolerance and microbial stimulation related to pet exposure might be of importance for immune development as well as pet-related allergen exposure. Beneficial effects with regard to adult asthma and allergy was further indicated for: (i) childhood cats – those with cats in childhood had less wheeze in areas with very high community prevalence of cats(27); (ii) growing up on a farm, which was related to less adult atopy and allergic rhinites, particularly in recent birth cohorts(35), and most convincing in subjects with a particular CD14C-159T polymorphism(36); (iii) and early exposure to other children, subjects from large families (particularly with brothers) or single children who had attended day care had less adult atopy(24) and less asthma(39) at all ages(28). These studies showing protective effects of early life exposures on adult allergy and allergic disease have contributed to further development of the hygiene hypothesis and further understanding of the mechanisms relating microbial stimulation to allergy. Harmful effects of early life factors on adult asthma and lung function have been indicated for severe respiratory infections early in life, which was associated with increased asthma(57) at all ages(28), lower lung function(58) and higher bronchial responsiveness(59), for parental smoking, which was related to more respiratory symptoms and lower lung function(44), and for lower birthweight, which was related to more asthma(60) and lower lung function in younger generations(66). Thus, the ECRHS has contributed to show both how early life environment may affect allergy development permanently in a beneficial way, and how childhood risk factors may have permanent adverse influence on adult respiratory health.

Methodological considerations

The main methodological problems in the study of early life environment and adult respiratory health are related to recall. Subjects may remember erroneously in two qualitatively different ways: they may remember inaccurately or they may remember systematically wrong. Differential recall bias, where the occurrence of the outcome may influence recall or reporting of the exposure, is a potential problem in retrospective studies. For example, the memory about one's mother having smoked during childhood could be influenced by having asthma and being more annoyed by the mother's smoking. This kind of bias may produce spurious associations. In the ECRHS, with a large sample size and information about several outcome measures, this problem can be addressed by, i.e. analysing the association of maternal smoking with lung function while excluding symptomatic subjects or subjects with ever asthma. Analysis of reliability in adult reporting of childhood pets showed that the reporting error was not associated with asthma or atopy; thus, differential recall bias was not indicated(23).

Scarce information is a major problem in studying effects of childhood factors on adult health. For example, one may expect that people remember whether their mother smoked or not when they were children, but not how many cigarettes a day and for how many years she smoked. This problem can to some degree be addressed by the large sample size of the ECRHS and by giving priority to power in analysis [i.e. use number of respiratory symptoms as an outcome variable rather than individual symptoms (67)], but it will mainly have to be dealt with in the interpretation of results.

The childhood environmental factors we have information about are closely interrelated, and confounding by unknown factors cannot be ruled out. For instance, large families more often keep pets and have smoking fathers, but there may be other characteristics of large families that we are not aware of and that may be important for asthma and allergy. It is evident that subjects cannot be randomised with regard to family size, maternal smoking or day care; thus, information about such factors will to some extent always be uncertain even in the best designed prospective cohort studies. Advice to the population about several important environmental factors therefore has to be based on cautious interpretation and comparison of several sources of imperfect evidence.

One major advantage that makes the ECRHS a substantial contributor within this field is the multi-cultural setting, together with the large sample size. The multi-cultural design provides an opportunity to separate biological effects that are homogeneous between centres from sociocultural effects that may vary between centres. For instance, it seems plausible that the biological effects of passive smoking on lung development is similar across Europe, while a sociocultural mechanism, like i.e. smoking father's spending more time with their sons, is likely to vary between centres. It is also valuable to see the results for each centre together with the results of all centres combined to understand the limitations of individual studies. The large sample size allows for analysis of differences between subgroups, like men and women, studies of only never-smokers, etc.

Another considerable advantage is that the ECRHS has information about adult outcome measures in a field where most knowledge comes from research on children.

Finally, analysis of the ECRHS is important to provide balanced information in areas where publication bias may be a problem. In the analysis of a multi-centre study, there is no room for publication of centres with ‘popular’ results only, and with the magnitude of the involvement in the ECRHS, there is an obligation to publish also complex or ‘unpopular’ findings.

For instance, the findings that dog keeping early in life was associated with less atopy contributed to the development of a new concept that contact with animals early in life might promote a non-allergic immunological maturation (24). This interpretation was possible, because the consistency of the results across centres and the large sample size gave confidence in the results, even if these were unsuspected and opposite to current understanding. However, protective effects of childhood pets was a popular finding, and it seems plausible that researchers, during a subsequent period of excitement, may have paid particular attention to supporting evidence. Further analysis of the ECRHS proved particularly useful in this context to show also the complexity of the area and the less popular findings: i.e. more respiratory symptoms in long-term dog keepers (27, 28).

Discussion of results

The influence of childhood pets on immunological and pulmonary development appears to be extremely complex, varying with the type of pet (24, 27), host factors (24, 27, 36) and community factors (27, 35). For practical purposes, childhood exposure to pets appears to be mostly bad with regard to asthma and respiratory symptoms, and possibly good with regard to allergy. The associations were relatively weak, this may be because of inconsistencies in a large multi-centre study, but may also be true as the ECRHS have the size and reputation to publish weak effects and thereby avoid publication bias. The results appear to be influenced by some selective avoidance of cats in asthmatics(31), and some non-differential misclassification bias attenuating associations(23)– these two types of errors would outweigh each other so that published effects of cats appears to be fairly correct, and the effects of dogs possibly slightly underestimated.

Could we avoid asthma by throwing out children's pets? Based on the published estimates, the protective effect of childhood dogs on adult hay fever (27) correspond to a population attributable risk of −5.5%; thus, childhood exposure to dogs may possibly reduce hay fever risk in the population by 5%–6%. On the other hand, counting together harmful effects of cats, dogs and birds on adult wheeze gives a population attributable risk of 15%, which is larger than, i.e. the 9% population attributable risk calculated for occupational factors in the ECRHS (68). Specifically addressing childhood cats and adult wheeze (27), 7% of adult wheeze may be attributed to childhood cat exposure in centres with low (<40%) community prevalence of cats (Germany, Estonia, Netherlands, Spain), in centres with medium high community prevalence of cats; 4% of adult wheeze may be attributed to childhood cats, while in centres with very high cat prevalence (Belgium, New Zealand, Melbourne Australia, Portland USA), childhood cats appeared to prevent 11% of adult wheeze. In conclusion, new knowledge has proven that the old general advice about avoiding pets in order to avoid allergy can no longer be defended. However, current scientific status does not support any general advice to the population at the moment. The individual advice to subjects with allergic symptoms related to a pet should, however, be as before – avoid pets. In the future, it is possible that genetic testing may give a basis for individual advice (36). This may be of great interest for individual subjects, but will most likely not have major importance for the prevalence of asthma and allergy on a population basis.

Growing up on a farm appears to be beneficial with regard to allergy development, but dubious with regard to asthma (35). The contribution of the ECRHS was to present weaker estimates than in other studies where publication bias may have favoured strong findings (35), effects that were consistent between culturally and geographically different centres (35), stronger effects in recent generations (35), and stronger effects in specific genetic subtypes (36). This supports the interpretation that microbial stimulation from a farm environment is beneficial for a non-allergic immune development in subjects with particular genetic features. The findings thus have increased our understanding of allergy, but have limited practical importance in the Western world, where few children grow up on farms.

The findings concerning family size also have limited public health importance, as family size is unlikely to be determined by relatively weak effects on atopy and asthma (28, 39). The effects of day care are also relatively weak and complex [day care was associated with less atopy (39) and asthma (28), but more respiratory symptoms(39)] and mostly concern the smaller group of single children (39) (10% in the ECRHS cohort); thus, the public health importance is limited. During the period when the ECRHS participants were born, from 1945 to 1972, sibling size decreased from medium 2.6 to medium 1.9 siblings per family, while day care attendance increased from 37% to 68%; the setting for contact with other children is to some extent changing from family to day care.

The findings on exposure to children as related to asthma and allergy are of great importance for increased understanding of the patophysiology behind allergy and asthma. ‘Family size’ is an even more complex ‘exposure’ than pets, and might reflect infectious agents and allergens, physical activity, selective fertility (the choice of having more children could be determined by having a severely allergic child), maternal immune stimulation because of repeated pregnancies, etc. The ECRHS contributed by showing that the inverse associations of atopy and hay fever with family size are amazingly consistent across centres (24, 39), suggesting a biological, rather than a sociocultural, explanation. The consistency in findings between family size and day care (39) suggests that contact with children is the key issue rather than other explanations. Thus, the results support the interpretation that microbial stimulation from contact with other children favours a non-allergic immune development. An interesting detail that may give a clue to new understanding of the allergic pathways was that the effect of family size on hay fever did not appear to be mediated by atopy (39).

The strong association of language group with allergic rhinitis (18) remains unexplained. Language group to some extent reflect genetic factors; this may be one possible explanation for the strong association. Several cultural factors are related to language group: i.e. smoking among young women was more common in (mostly English-speaking) countries with influence from US soldiers during the Second World War; religion is related to language group and also to various lifestyle factors, like smoking and diet, i.e. did women in Catholic countries not smoke during the post-war period; Mediterranean diet coincides to some extent with Latin languages in Europe, etc. Thus, there are unknown genetic and/or cultural factors that appear to be of considerable importance for allergy.

Parental smoking give permanently reduced lung function and increased respiratory symptoms in offspring (44). This was convincing in analyses of the ECRHS, even though based on information obtained from adults. Smoking among women increased during the study period, and, in the English-speaking centres where maternal smoking was common, parental smoking accounted for 14% of adult wheeze. The lung function deficit related to parental smoking was comparable with that of 6–9 years of active smoking. When also considering the fact that subjects with parents who smoked were more often smokers and smoked heavier (44), the impact of parental smoking on adult lung health must be considered to be substantial. Thus, early life exposure to parental smoking is an important risk factor, where prevention is likely to improve lung health in the population considerably.

Subjects with severe respiratory infections in early childhood had considerably lower adult lung function (58) and markedly increased asthma incidence at all ages (28), as well as slightly increased bronchial responsiveness when adult (59). It is likely that lower airways infections during sensitive time windows in pulmonary development may have permanent adverse effects on the lungs, leaving a potential for intervention. On the other hand, it is likely that host factors contribute to determine susceptibility to and severity of lower airways infections. The potential for prevention is thus difficult to estimate, however, it seems likely that vaccination against microbes causing common respiratory infections might prove beneficial for long-term respiratory health.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References

Analyses of the ECRHS have contributed substantially to increased knowledge about the early life origins of allergy and obstructive pulmonary disease. The multi-centre structure of the study, the large sample size, the extensive information about each participant, follow-up until age of 56 years and the study's reputation have given the basis for convincing conclusions, and made possible the publication of weak and unsuspected findings in spite of the problems related to adult recall of childhood events.

Data from the ECRHS strongly suggests that early life environment have permanent influence on allergy, asthma and lung function, in accordance with the Barker hypothesis. Several analyses from the study support the hygiene hypothesis that microbial stimulation early in life may in some cases favour a non-allergic immunological development. ECRHS researchers has developed this concept further by suggesting that pets also may provide beneficial microbial stimulation, and thereby contributed to new thinking around the possible roles of animals in allergy. There appears to be more to pets than allergic sensitisation; microbial stimulation, development of tolerance and non-allergic inflammation must also be considered. Atopy may be of less importance than previously considered, as the effects of microbial stimulation on allergic disease do not appear to be mediated by atopy. Analyses of the ECRHS also give some pieces of evidence that the age-window of sensitivity to harmful exposures may differ between men and women; one should be aware of possible gender differences also when investigating the early life origins of respiratory health.

From a practical healthcare perspective, improved care for mother and child is likely to improve general respiratory health in the population. Analyses of the ECRHS suggest that avoiding parental smoking would improve offspring's pulmonary health considerably. Vaccination against common airways pathogens in children might possibly improve respiratory health on a population basis. Improved microbial stimulation should, from a theoretic point of view, reduce allergy in the population; however, we do not know precisely what would be beneficial microbial stimulation, while the harmful effects of infections are well known. Thus, current knowledge does not yet permit practical advice in this area. Specifically, the current scientific basis does at the moment not allow any general advice to the population concerning pet keeping and prophylaxis of asthma and allergy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Design
  6. Assessment of early life factors
  7. Assessment of adult respiratory health
  8. Results
  9. Discussion
  10. Conclusion
  11. References