a Unit of Paediatric and Perinatal Epidemiology, Institute of Child Health,
b Department of Social Medicine,
c Institute of Child Health, Royal Hospital for Sick Children, University of Bristol, Bristol, UK.
d Department of Public Health Sciences, St George's Hospital Medical School, London.
Dr Andrea Sherriff, Unit of Paediatric and Perinatal Epidemiology, University of Bristol, 24 Tyndall Avenue, Bristol BS8 1TQ, UK. E-mail: Andrea.Sherriff{at}bris.ac.uk
Abstract
Background There is a paucity of detailed longitudinal data on wheeze in early childhood. Not all children who wheeze in early infancy will continue to wheeze into childhood and beyond. This study aims to investigate possible risk factors for different patterns of wheeze in the pre-school years.
Subjects and Methods Study participants were part of the Avon Longitudinal Study of Parents and Children (ALSPAC). Maternal reports of child wheeze between birth and 6 months and again between 30 and 42 months were gathered prospectively. Children were categorized into early wheeze, persistent wheeze or late onset wheeze. A large number of risk factors were assessed for each wheezing phenotype using multivariable logistic regression models.
Results Over 70% of children who wheezed in the first 6 months did not wheeze 3 years later. Wheezing between 06 months was independently associated with the presence of older siblings, male sex, delivery between April and December, bottle feeding, young maternal age, prenatal tobacco smoke exposure, atopy and parental history of asthma. From within this group of early wheezers, risk factors for wheeze that persisted beyond 6 months included pre-term delivery, young maternal age, living in rented local authority housing, atopy and a maternal (not paternal) history of asthma. Atopy and a family history of asthma emerged as the main predictors of wheeze that developed after 6 months of age.
Conclusion It is clear that a number of wheezing syndromes exist by 3 years, albeit with some degree of overlap. Detailed follow-up of this cohort is underway to determine whether risk factor associations determined in the first 3
years have long-term significance for the clinical entity termed asthma.
Keywords ALSPAC, infant wheeze, risk factors
Accepted 2 March 2001
Previously, studies of the determinants of childhood wheeze have failed to elucidate fully the pathogenesis of the condition or its association with asthma. Recently, however, studies by Martinez et al.1 and Rusconi et al.2 have recognized that childhood wheeze is a heterogeneous condition which does not necessarily develop into the clinical entity termed asthma.
Not all children who wheeze in early infancy will continue to wheeze into childhood and beyond. Some wheezing, particularly during viral respiratory infections, is thought to result from small airways or abnormal airway dynamics in very young children. Over time, the majority of these children stop wheezing as their airways grow, however for other children the occurrence of early wheeze is a predictor of a chronic condition that persists beyond childhood.
Due to the paucity of detailed longitudinal data on wheeze in early childhood, there remains much to be learned of the natural history of wheeze, including the age at which different wheezing conditions (phenotypes) can be distinguished. Martinez et al.1 and Rusconi et al.2 both examined children at age 6, however the lower age cut-off points differed between the studies (3 years for Martinez et al. and 2 years for Rusconi et al.), leaving unanswered the question as to the age at which separate wheezing conditions can be distinguished. By distinguishing wheezing phenotypes earlier in childhood, there is a greater chance that intervention programmes set up to alter the natural history of asthma would be successful. In addition to this, low power in the study by Martinez et al. may have resulted in subtle associations being overlooked, and the retrospective element of the Rusconi et al. study may have led to significant recall bias.
Thus, the motivation for the present study was:
(1) to overcome the methodological problems of the previous studies by using longitudinal data that was gathered prospectively on a large contemporary population cohort based in the UK (The Avon Longitudinal Study of Parents And Children [ALSPAC]).
(2) To challenge the conventional cut-off points used to define wheezing phenotypes, by looking at wheezing patterns within the first 31/2 years.
(3) To address clinically important questions regarding wheezing conditions within the first 31/2 years.
Using data gathered on children participating in ALSPAC, at two appropriately spaced study periods: 06 months and 3042 months, the following research questions were posed:
(a) Which risk factors are associated with wheezing in the first 6 months after birth? (early infant wheeze),
(b) Within the group of children who wheezed in the first 6 months after birth, which risk factors are associated with wheeze persisting into the period 3042 months? (persistent wheeze),
(c) Within the group of children who did not wheeze in the first 6 months, which risk factors are associated with wheeze reported during the period 3042 months? (late onset wheeze).
Methods
Study sample
The Avon Longitudinal Study of Parents and Children (ALSPAC)3 is a prospective study of 14 541 pregnancies that resulted in 13 971 live births that survived to one year. Women were enrolled as early in pregnancy as possible on the basis of an expected date of delivery between 1 April 1991 and 31 December 1992 and place of residence within the three Bristol-based health districts of the former county of Avon, UK. It was estimated that 8590% of eligible mothers were enrolled in the study. Data were collected from questionnaires completed by the parents, medical records and biological samples. The study mothers completed four questionnaires during pregnancy. Questionnaires relating to the child were administered at 1, 6, 15, 18, 24, 30 months and 6-monthly intervals, and to the study mothers and their partners on an annual basis.
Data collection
Data on wheeze were collected from child-based questionnaires administered at 6 and 42 months postnatally. Identical health questions for both study periods were asked. The present analysis considered responses to the question: Since he/she was born (at 6 months)/In the last 12 months (at 42 months) has he/she had any periods when there was wheezing with whistling on his/her chest when he/she breathed? for the two periods: birth6 months and 3042 months.
A total of 10 409 child-based questionnaires were returned by study mothers whose children were aged between 4 and 8 months. Over 80% (N = 8594) of these mothers returned questionnaires once more when their children were 31/2 years of age.
The study and comparison groups
To study risk factors for wheeze in the first 6 months (research question (a)), all children who wheezed 06 months were compared to all children who did not wheeze 06 months (N = 1613 versus 6981). To study risk factors for wheeze that developed in the first 6 months and persisted beyond 30 months (research question (b)), those children who wheezed in the period 06 months but had stopped by 3042 months were compared to those children who wheezed during both study periods (N = 433 versus 1180). Finally to study risk factors for wheeze that developed after 6 months of age (research question (c)), children who did not wheeze between 06 months but did so between 3042 months were compared to children who had never wheezed during either study period (N = 729 versus 6252).
Putative risk factors
Putative risk factors for childhood wheeze were grouped into a number of categories prior to analysis based on previously reported associations with wheeze or theoretical prior hypotheses. The groups were perinatal, social, parental, allergic disposition and environmental risk factors. The categories associated with each of the putative risk factors are detailed in Table 1.
|
Social risk factors
Housing tenure, marital status, and highest maternal educational achievement were ascertained from questionnaires administered during pregnancy. Maternal age at delivery was obtained from obstetric records.
Parental risk factors
Maternal history of asthma was ascertained during pregnancy. A history of asthma in the partner of the study mother was obtained from one of two sources: from the questionnaire sent via the study mother to her current partner, if it had been returned; otherwise the study mother's own report was used. Concordance between maternal report of her partner's asthma history and partner's own report was around 80% ( = 0.6).
Allergic disposition (atopic risk factors)
At 6 and 42 months (31/2 years) mothers were asked whether their baby/young child had a rash in the joints and creases of their body and whether there was an itchy, dry, oozing or crusted rash on the face, forearm or shins. Responses from these questions gave a proxy measure of allergic disposition of the child at 6 and at 42 months of age. From these an atopic score was derived for each child. The score ranged from 0 to 4, with a score of 0 indicating that the child had no reports of any rash during either study period whereas a score of 4 indicated the child was reported to have both kinds of rash during both study periods.
Environmental risk factors
A combined smoking variable was derived that took into account both prenatal exposure to tobacco smoke through maternal smoking during pregnancy, and postnatal exposure from environmental tobacco smoke (ETS) at the weekend. Prenatal smoking was determined at approximately 32 weeks gestation. Postnatal ETS exposure was assessed when the child was 6 months old using a measure of weekend exposure. Weekend exposure was used as it was judged to be the best representation of composite tobacco smoke exposure of an individual subject. Ownership of a domestic pet was determined from questionnaires returned around the study child's first birthday.
Statistical Analyses
Statistical analyses were carried out using SPSS for Windows (version 9.0.0). The three research questions (a)(c) above were addressed separately in the analyses, although identical methodology was adopted for each. First, putative risk factors for infant wheeze (early, persistent and late onset) were analysed individually using univariable logistic regression models, and statistical significance was based on the likelihood ratio statistic (for heterogeneity).
The multivariable analyses of infant wheeze were carried out in several stages using Backwards Method of elimination (P > 0.2 from the likelihood ratio test for exclusion). Putative risk factors for infant wheeze (early, persistent, late onset) were first analysed simultaneously within each of the risk factor groups in an effort to reduce colinearity and minimize the chance of producing misleading parameter estimates. Statistically significant risk factors from within the perinatal and social groups were then combined and offered simultaneously to a further set of models of infant wheeze. Similarly, risk factors from within the parental, the allergic disposition and the environmental exposure groups were then considered together. Those risk factors remaining independently statistically significant were then combined to produce a final model for each infant wheeze outcome. Odds ratios (OR) and 95% CI resulting from the final models were adjusted for all other factors remaining in those models.
Univariable and multivariable analyses were carried out using all available data for each of the analyses. The numbers of children included in each analysis are given in the Tables. To determine whether there was any selection bias using this technique, all analyses were repeated using data from those children who had complete information on all risk factors (N = 7224/8595 = 85%). The outcomes of the analyses using both techniques were the same. Thus, the inferences from the analyses were unchanged.
Results
Descriptive data are presented in Table 1. Although it was not the purpose of this study to compare all wheezing groups with the never wheezed group, Table 1
can be used to do so. From Table 1
, 18.8% (N = 1613) of children were reported to wheeze in the first 6 months (early infant wheeze), 5% (N = 433) of the children wheezed during both study periods (persistent wheeze), and 8.5% (N = 729) of the study population were reported to wheeze between 30 and 42 months but not between birth and 6 months (late onset wheeze). Only 27% (433/1613) of children who were reported to wheeze between birth and 6 months were also reported to be wheezing 2 years later during the period 3042 months.
Table 2 presents results of the unadjusted analyses where individual associations between the putative risk factors and early infant wheeze, persistent wheeze and late onset wheeze, respectively, were assessed. Tables 3, 4 and 5
give results of the multivariable modelling of the three infant wheeze outcomes and on which the following section is based. The sample sizes for each of the analyses are given in Tables 3, 4 and 5
.
|
|
|
|
Children with atopy or a family history of asthma (both mother and her partner) were more likely to develop early wheeze. With every unit increase in rash score, the likelihood of early wheeze increased by 20%. Contact with cats or dogs in the child's first year and prenatal exposure to tobacco were also independently significant risk factors for early wheeze.
Persistent wheeze: Comparing those who wheezed in the first 6 months and stopped with those who persisted wheezing after 6 months
From Table 4 children who were reported to wheeze in the first 6 months after birth were more likely to continue wheezing (rather than stop) if they had been delivered pre-term, if there were two or more older siblings in the household and if they were living in council rented accommodation (rather than mortgaged or owned property). Unlike early infant wheeze this effect was independent of tobacco smoke exposure and of all other factors remaining in the final model. There was some indication that male sex and bottle feeding in the first 6 months were risk factors for persistent wheeze, however these were of borderline significance.
For every unit increase in skin rash score, the risk of continuing to wheeze after 6 months rather than stop increased by 40%. In addition, almost double the proportion of children had persistent wheeze if their mother had had a history of asthma. However, the risk of wheeze persisting, having started in the first 6 months, was not significantly increased if the study mother's partner had a history of asthma. Maternal age at delivery was associated with persistent wheeze, although it was only children born to mothers in the age group 2024 years who were more likely to persist wheezing beyond 30 months.
Late onset wheeze: Comparing those children who did not wheeze in the first 6 months but did so in the period 3042 months with those children who did not wheeze during either period
Overall there were fewer risk factors associated with late onset wheeze than early infant wheeze or persistent wheeze. Within the group of children who did not wheeze at 06 months, children of male gender or of low birthweight (<2500 g) were more likely to develop wheeze after the age of 6 months when compared with females or children of normal birthweight. Late onset wheeze was more common in children with a family history of asthma, both in the study mother and in her partner. In both cases, children with familial asthma were almost twice as likely to have late onset wheeze than children with no family history. High rash scores were also strongly associated with late onset wheeze. The likelihood of developing wheeze after 6 months of age (if no wheeze was reported between 06 months) increased by 50% for every unit increase in rash score.
Discussion
This study has demonstrated that different patterns of wheeze in children up to the age of 31/2 years are associated with different risk factors, albeit with some degree of overlap. The results from the present study could not be directly compared to those of previous studies as the research questions, and hence, the comparison groups, differed in detail. The questions posed in this study were based on clinically important issues such as why some children wheeze in the first 6 months, yet others do not develop wheeze until after 6 months of age; and why some children who wheeze in the first 6 months stop yet others continue to do so. The risk factors that emerged for wheezing in the first 6 months (comparison group: those who never wheezed in the first 6 months) were comparable to risk factors associated with early respiratory infections. These included the presence of older siblings, bottle feeding rather than breastfeeding, contact with cats and dogs and exposure to tobacco smoke, particularly prenatal exposure. Although atopy and familial asthma were important risk factors for early wheeze, they also significantly increased the likelihood that early wheezing would persist beyond 6 months. Additionally, within the group of early wheezers, those children who continued to wheeze beyond 6 months (rather than stop) were more likely to be male, pre-term deliveries, have older siblings and live in local authority rented accommodation. In contrast, there were few perinatal, social or environmental risk factors associated with late onset wheeze. Instead, atopy and a family history of asthma emerged as the primary risk factors for wheeze that did not develop until after 6 months of age (comparison group: never wheezed). These findings support the hypothesis that childhood wheeze is a heterogeneous condition which may lead to a number of different wheezing outcomes in the future.
The population of Avon is broadly representative of the whole of Great Britain (1991 census)3 and due to the prospective nature of the study, recall bias has been limited.
Of the 8594 children in this sample, 7224 had complete data for multivariable analyses (85%). Those children with missing data (N = 1370) were more likely to wheeze in the study period than those children with complete data (Missing group: 33% wheezed versus Complete data: 26% wheezed; P < 0.0001). There were no significant differences between these groups with respect to gender, maternal parity, season of birth, atopy, familial history of asthma or ownership of domestic pets. However, compared to children with complete data, those children with missing data were more likely to be of low birthweight, pre-term, bottle-fed, living in local authority housing and have a higher exposure to tobacco smoke, both in the prenatal and postnatal period. In addition, their mothers were more likely to be younger, not married with lower educational achievements. Coupled with the fact that there was a higher proportion of wheezers in the missing data group and the factors differentiating the two groups were all potential risk factors for wheeze in childhood, no significant changes to the direction of the reported effects would have been expected.
It is well documented that overcrowding in the home is positively associated with early viral respiratory infections.48 Rusconi et al.2 did not find any association between sibship size and persistent wheeze but found a positive association with early transient wheeze and an inverse relationship with wheeze of late onset, consistent with the so-called hygiene hypothesis. Although we found that increasing maternal parity was associated with wheezing in the first 6 months, there was no protective effect of older siblings against wheeze of late onset when compared to the never wheezed group. There may have been some misclassification of children into the late onset group who would have been more sensibly classified as early transient wheeze (i.e. those children who start wheezing in the period 630 months). This misclassification may have attenuated the effect of maternal parity in this group.
Rusconi et al. found an independent protective effect of prolonged breast feeding against early transient wheeze, but not persistent wheeze or wheeze of late onset.2 However, Oddy et al. found that the age of introduction of non-breast milk was inversely associated with several asthma outcomes at 6 years of age.9 In the ALSPAC cohort, infants who were breastfed in the first 6 months were less likely to wheeze in the first 6 months than bottle-fed infants. However, there was no evidence for any long-term protection conferred by breast milk on wheeze. Differences in the measurement of early feeding behaviour and other methodological issues may explain different study outcomes. Alternatively children in the aforementioned studies may have been too young for the protective effect of breastfeeding to manifest itself. One very long-term study of atopic diseases has reported that the positive effects of prolonged breastfeeding on respiratory allergy are not manifested until 17 years of age.10 Further investigation of early feeding behaviour and detailed follow-up of wheezing patterns throughout the childhood and adolescence of our study sample may help to elucidate the putative protective effects of breast milk against atopic disease.
There was a clear association between early infant wheeze and young maternal age (<25 years), particularly in teenage mothers (<20 years). There is some evidence to show that both metabolic and socioeconomic factors are associated with poorer pregnancy outcome for teenage mothers.1112 Our result is independent of a number of confounding social and perinatal factors, yet we cannot rule out the possibility that young maternal age may be a proxy for some other unknown social indicator.
Early infant wheeze and wheeze of late onset were more common if the study mother had asthma and independently if her partner had reported asthma. The propensity to wheeze in childhood is closely related to familial asthma history and asthma is known to aggregate in families. A number of published genetic studies, have confirmed earlier epidemiological findings of a preferential maternal transmission,1315 however others have failed to replicate these findings.1617 The strength of the effect of paternal asthma on both early and late onset childhood wheeze in this study is similar to that of maternal asthma, a finding that is increasingly common in studies where partners asthma status is ascertained directly.18
The finding in this study that children as young as 31/2 years of age are demonstrating a differential association between wheezing patterns and markers of atopy (i.e. skin rashes) is new and hence requires further research on this group of children in terms of objective biological markers of atopy and asthma (skin prick tests, total IgE levels and lung function tests) and in terms of more detailed symptoms as the children get older.
Children born to mothers who smoked during pregnancy are on average smaller at birth for gestational age, more at risk of being delivered pre-term and have higher perinatal mortality for their birthweight.1921 There is evidence to show that their lung function is lower compared with children born to non-smokers.1,2123 In a previously published ALSPAC study,24 as in the present study, exclusive prenatal tobacco smoke exposure but not exclusive ETS was a risk factor for early wheeze. There was no evidence that tobacco smoke exposure provoked wheezing in later onset or persistent wheezing groups. In one study, postnatal smoking was found to be less associated with the incidence of bronchitis in young children than smoking during pregnancy.25 This may reflect the detrimental effect of tobacco smoke on lung growth in utero which others have shown to have a benign prognosis later in childhood.
Conclusion
In this study of early childhood wheeze, it is clear that a number of wheezing syndromes exist with some degree of overlap. The majority of children who wheeze in the first 6 months are no longer wheezing 3 years later. This group is characterized by a number of risk factors associated with early infection (maternal parity, non-breastfed, season of birth) that precipitates wheezing in children with small airways or reduced airway calibre (tobacco smoke exposure, maternal age). Familial history of asthma and strong associations with atopic markers define the late onset wheeze group, whereas the children who wheeze early and continue wheezing after 30 months (persistent wheeze) have risk factors common to both groups. This may reflect a predisposition to early wheeze that is subsequently predictive of a more chronic condition later in childhood. Distinguishing between wheezing phenotypes as early in childhood as possible is crucial to the success of any future public health intervention programmes set up to alter the natural history of asthma.
KEY MESSAGES
|
Acknowledgments
We are extremely grateful to all the mothers who took part and to the midwives for their co-operation and help in recruitment. The whole ALSPAC study team comprises interviewers, computer technicians, laboratory technicians, clerical workers, research scientists, volunteers and managers who continue to make the study possible. This study could not have been undertaken without the financial support of the Medical Research Council, the Wellcome Trust, the UK Department of Health, the Department of the Environment, and DfEE, the National Institutes of Health, a variety of medical research charities and commercial companies. The ALSPAC study is part of the WHO initiated European Longitudinal Study of Pregnancy & Childhood.
References
1
Martinez FD, Wright AL, Taussig LM et al. Asthma and wheezing in the first six years of life. N Engl J Med 1995;332:13338.
2
Rusconi F, Galassi C, Corbo GM et al. Risk factors for early, persistent and late-onset wheezing in young children. Am J Respir Crit Care Med 1999;160:161722.
3 ALSPAC web page: http://www.ich.bris.ac.uk/alspac.html
4 Nystad W, Skrondal A, Magnus P. Day care attendance, recurrent respiratory tract infections and asthma. Int J Epidemiol 1999;28:88287.[Abstract]
5
Celedon JC, Litonjua AA, Weiss ST, Gold DR. Day care attendance in the first year of life and illnesses in the upper and lower respiratory tract in children with a familial history of atopy. Pediatrics 1999; 104:495500.
6 Flores Hernandez S, Reyes Morales H, Perez Cuevas R, Guiscafre Gallardo H. The day care center as a risk factor for acute respiratory infections. Arch Med Res 1999;30:21623.[ISI][Medline]
7 Leach AJ. Otitis media in Australian Aboriginal children: an overview. Int J Pediatr Otorhinolaryngol 1999;49(Suppl.1):S17378.[ISI][Medline]
8 Rahman MM, Rahman AM. Prevalence of acute respiratory tract infection and its risk in under five children. Bangladesh Med Res Counc Bull 1997;23:4750.[Medline]
9
Oddy WH, Holt PG, Sly PD et al. Association between breast feeding and asthma in 6 year old children: findings of a prospective birth cohort study. Br Med J 1999;319:81519.
10 Saarinen UM, Kajosaari M. Breast feeding as prophylaxis against atopic disease: prospective follow-up study until 17 years old. Lancet 1995;346:106569.[ISI][Medline]
11 Schwartz J, Gold D, Dockery DW et al. Predictors of asthma and persistent wheeze in a national sample of children in the United States. Am Rev Resp Dis 1990;142:55562.[ISI][Medline]
12 Martinez FD, Wright AL, Holberg CJ, Morgan WJ, Taussig LM. Maternal age as a risk factor for wheezing lower respiratory illnesses in the first year of life. Am Epidemiol 1992;136:125868.[ISI]
13 Shirakawa T, Li A, Dubowitz M et al. Association between atopy and variants of the ß subunit of high affinity IgE receptor. Nature Genet 1994;7:12530.[ISI][Medline]
14
Hill MR, James Al, Ryan G et al. Fc RI-ß polymorphism and risk of atopy in a general population sample. Br Med J 1995;311:77679.
15 Daniels SE, Bhattacharrya S, James A et al. A genome wide search for quantitative trait loci underlying asthma. Nature 1996;383:24750.[ISI][Medline]
16 Hizawa N, Yamaguchi E, Furuya K et al. Association between high serum total IgE levels and D11S97 on chromosome 11q13 in Japanese subjects. J Med Genet 1995;32:36369.[Abstract]
17 Doull IJM. Maternal inheritance of atopy? Clin Exp Allergy 1996; 26:61315.[ISI][Medline]
18
The European Community Respiratory Health Survey Group. Genes for asthma? An analysis of the European Community Respiratory Health Survey. Am J Respir Crit Care Med 1997;156:177380.
19 Wilcox AJ. Birthweight and perinatal mortality: the effect of maternal smoking. Am J Epidemiol 1993;137:1098104.[Abstract]
20 Lieberman E, Gremy I, Lang JM, Cohen AP. Low birthweight at term and the timing of fetal exposure to maternal smoking. Am J Public Health 1994;84:112731.[Abstract]
21 Doull IJM, Holgate ST. Asthma: early predisposing factors. Br Med Bull 1997;53:7180.[Abstract]
22 Young S, Lesouef PN, Geelhoed GC, Strick SM, Turner KS, Landau LI. The influence of a family history of asthma and parental smoking on airway responsiveness in early infancy. N Engl J Med 1991; 324:116873.[Abstract]
23
Cook DG, Strachan DP, Carey IM. Parental smoking and spirometric indices in children. Thorax 1998;53:88493.
24 Lux AL, Henderson AJ, Pocock SJ, ALSPAC Study Team. Wheeze associated with prenatal tobacco smoke exposure: a prospective, longitudinal study. Arch Dis Child In press.
25 Taylor B, Wadsworth J. Maternal smoking during pregnancy and lower respiratory tract illnesses in early life. Arch Dis Child 1987; 62:78691.[Abstract]