- Email: [email protected]
The role of domestic factors and day-care attendance on lung function of primary school children
RESPIRATORY
MEDICINE
(1998)
92,
928-935
The role of domestic factors and day-care attendance on lung function of primary school children K. DEMISSIE*, P. ERNST*, L. JOSEPH+ AND M. R. BECKLAKE* “Department of Epidemiology and Biostatistics, Respiratory Epidemiology Unit, McGill University, 1110 Pine Avenue West, Montreal, Quebec, Canada H3A 1A3 ‘Departmenf of Epidemiology and Biostaitistics,McGill University, 1020 Pine Avenue West, Monfreal, Quebec, Canada H3A IA2 The results of studies examining the relationship of domestic factors to lung function are contradictory. We therefore examined the independent effects of maternal smoking during pregnancy, exposure to environmental tobacco smoke (ETS), the presence of a cat, type of heating and cooking used in the home and day-care attendance on lung function after controlling for socioeconomic status (SES). Nine hundred and eighty-nine children from 18 Montreal schools were studied between April 1990 and November 1992. Information on the child’s health and exposure to domestic factors was collected by questionnaire. Spirometry was performed at school. The data were analysed by multiple linear regression with percent predicted FEV,, FVC, and FEV,/FVC as dependent variables. In the overall sample (both sexescombined), cat in the home (regression coefficient, p= - 1.15, 95% confidence interval, CI: - 2.26- 0.05) and electric baseboard units (J= - 1.26, 95% CI: - 2.39- - 0.13) were independently associated with a lower FEVJFVC, while day-care attendance (j?= - 2.05, 95% CI: - 3.71- - 0.40) significantly reduced FEV,. Household ETS was significantly associated with increasing level of FVC (p=2.86,95% CI: +0.55 to +5.17). In boys but not girls, household ETS @= - 2.13, 95% CI: - 4.07- - 0.19) and the presence of a cat u= - 2.19, 95% CI: - 3.94- - 0.45) were associated with lower FEV,/FVC. By contrast, day-care attendance was associated with lower FEV, (p= - 2.92, 95% CI: - 5.27- - 0.56) and FEV,/FVC @= - 1.53, 95% CI: - 2.73- - 0.33) in girls only. In conclusion, the results provide evidence that domestic factors and day-care attendance primarily affected airway caliber and gender differences were apparent in the effects of these factors. RESPIR. MED.
(1998)
92,
928-935
Introduction Various indoor and outdoor pollutants have been found to be associated with lower respiratory diseasesof infants and children. Children in Western countries spend more than 80% of their time indoors (1,2). Furthermore, the weight of evidence suggests indoor (as opposed to outdoor) pollutants to be more important as a risk factor for respiratory ill health in children (3). Concentrations of respirable particulate, nitrogen dioxide, and dust allergens are higher in homes where gas is used for cooking and in homes that contain pets (4-7). Received 21 January 1997, and accepted in revised form 26 February 1998. This work was supported by the Respiratory Health Network of Centers of Excellence and the Medical Research Council of Canada. Correspondence should be addressed to: K. Demissie, Department of Family Medicine, UMDNJ-Robert Wood Johnson Medical School, One Robert Wood Johnson Place-CN 19, New Brunswick, NJ 08903-0019, U.S.A. 0954-6111/98/070928+08
$12.0010
Several investigators have found an association between environmental tobacco smoke (ETS) and reduced lung function in the early postnatal period (8-15). Most of these studies were unable to assessthe independent contributions of prenatal and early postnatal exposure to tobacco smoke on lung function. Furthermore, information as to whether a deficit in lrtero or in the early postnatal period persists to childhood or is compensated by a catch-up growth has only been reported infrequently (8). Although the relationships of pet exposure and the development of sensitization and allergic diseaseshave been extensively investigated (16,17), data on the relationship of exposure to pets and lung function is limited. In a study among 1357 non-smoking children in the U.S.A., the presence of indoor pets was not found to be associated with a lower FEV, (18). The relationship between type of home heating and cooking fuel used in the home with children’s lung function levels is less consistent. Several studies of gas cooking and pulmonary function levels have found no association (1921), while Hasselblad et al. (22) and Ware et al. (23) found 0
1998
W. B. SAUNDERS
COMPANY
LTD
DOMESTICFACTORSAND
reduced FEV, among children living in homes with gas stoves. More recently, Hosein et al. (18) reported the use of gas stoves, the absence of air conditioning and the use of hot water heating to be associated with lower FEV,. The type of home heating, use of air conditioning, cooking fuel use and exposure to ETS are likely to differ according to socioeconomic status (SES). Most studies reporting on the effect of such indoor factors did not control for SES or used imprecise methods of assessing social class that are likely to have resulted in residual confounding. We therefore examined the independent effects of maternal smoking during pregnancy, postnatal exposure to ETS, presence of a cat in the home, type of home heating and cooking fuel used in the home, parental asthma, and day-care attendance on lung function in a study of school children after accounting for the possible effects of SES, gender, age, race, height and body mass index of the child.
Methods
LUNG FUNCTION
IN CHILDREN
929
Symptoms and diagnoses were defined as follows: ‘Does the child’s chest ever sound wheezy or whistling? (ever wheeze); ‘Does the child usually cough?’ (usual cough); ‘Has the child ever been diagnosed as having asthma?’ (history of asthma); ‘Does your child take any medications for breathing?’ (asthma treatment); ‘Does the child’s natural father suffer from asthma?‘/‘Does the child’s natural mother suffer from asthma?’ (asthma in a parent). LUNG FUNCTION In the school gymnasium, each subject’s age, gender, height, and weight were recorded, and the children were asked whether they had experienced a respiratory tract infection within the previous week. Spirometry was carried out sitting and with nose clips using two Collins 10-l water-sealed spirometers (Warren E. Collins, Braintree, MA, U.S.A.) according to current American Thoracic Society guidelines (25). The best FEV, and FVC from any acceptable flow-volume curve were used for analysis (26).
STUDY POPULATION
ASSESSMENT OF SOCIOECONOMIC
Eighteen schools were selected on the island of Montreal in order to represent a broad range of socioeconomic status (SES).To achieve this, all schools in the five school boards in central Montreal were ranked according to neighborhood average house values published by Statistics Canada (24). Within each school board, schools were selected from the upper, middle and lower ranges of neighborhood average house value. For each of the 18 schools so selected, three classesfrom each of grades one (5-7 years of age), three (8 and 9 years of age) and five (lo-13 years of age) were selected. If more than one class of a grade was available, we took the first in which the teacher agreed to participate and approached the teachers at their convenience or in the order suggested by the school principal. A total of 1274 children were recruited for the study. Comparison of schools selected (n=18) with those not selected (n=291) showed no meaningful difference with respect to neighborhood poverty, income and educational attainment (24).
Parental occupation was transformed into an SES score by coding the last occupation of the parents into the corresponding codes of the Canadian Classification and Dictionary of Occupations (27). These codes were then converted into SES scores (the highest score from either parent was retained for analysis) for the child, based on the income and education level for each occupation from the tables developed by Blishen et al. (28). The SES score was used to group the children into five categories. Children from unemployed families were grouped into category I, whereas categories II to V were formed using the quartiles of the SES score, category II with the lowest and category V with the highest score. respectively.
RESPIRATORY QUESTIONNAIRE A questionnaire about the child’s health and home environment was completed by parents (in 73% by biological mothers, 24% by biological fathers, and 3% by either a step or grandparent). Parents were asked about the current presence of pets (dogs, cats, birds, hamsters/mice/guinea pigs) in the home, the type of cooking and heating fuel used in the home, the number of people sleeping in the child’s bedroom (child shares bedroom) and whether the child had attended day-care. The child’s exposure to environmental tobacco smoke both in utero (maternal smoking during pregnancy) and postnatally was also assessed, and an attempt was made to quantify current smoking exposure by determining the number of smokers in the home and their average number of cigarettes smoked per day at home (household ETS).
STATISTICAL
STATUS
METHODS
The outcome variables were FEV,, FVC, and the ratio of FEV, to FVC (FEV,/FVC). The explanatory variables considered in this analysis were maternal smoking during pregnancy, household ETS, type of home heating and cooking fuel, current presence of pets, asthma in a parent, day-care attendance, gender, age, race, height and body mass index of the child. In children, it has been recommended not to use multiple regression on untransformed lung function data (29). To standardize the lung function values, FEV, and FVC were each transformed into a natural logarithmic scale (In) and regressed on In (standing height), In (age), and In (body mass index), gender and race. Constant variance and normality was obtained in the residuals. The antilogarithm of the residuals multiplied by 100 provided the percent predicted lung function values (using an internal standard value obtained among all children). The relationships of the percent predicted lung function values to the categorical independent variables were evaluated by multiple linear regression before and after accounting for the effects of
930
K. DEMISSIE ET AL.
potentially confounding variables. For FEV,/FVC, normality and homogeneity of variance could be obtained by the arc sine transformation. The equation for the calculation of percent predicted FEV,/FVC was obtained from this transformed variable. Multiple linear regression was used to evaluate the relationship between the independent variables and percent predicted FEV,/FVC after back transforming the ratio to the original scale.
Results A total of 1274 children were eligible to participate in the study from the 18 selected Montreal schools. Of this, the parents of 130 (10.2%) refused participation for their children, whereas a further 75 (59%) children did not return the questionnaire and consent form. Spirometry was not performed for a further 23 children because they were absent from school. Of the 1046 children who performed the spirometric test, 28 (2.7%) did not have acceptable spirometric data. The spirometry data of 28 (3%) children were lost after the test. One child (0.1%) was excluded because of severe asthmatic attack at the time of the test. There were no meaningful differences between participants and non participants as to the age of the child, sex and socioeconomic status as assessedby neighborhood census data. After the above exclusions, 989 (78%) children remained for analysis. Seven hundred and sixty-six (78%) of these children were Caucasian and ranged in age from 5 to 13 years, though the great majority (99.4%) were 6 to 12 years of age. Descriptive characteristics of the children whose parents responded to the initial questionnaire are presented in Table 1. The prevalence of wheezing was quite low in comparison to published results in similar populations. This is likely to be due to the unsatisfactory translation of the term wheezing into French and has been described previously in studies carried out in the province of Quebec (30). Size measurements and lung function levels (crude and standardized) as well as demographic variables are displayed in Table 2. The effect of different exposures in the home on lung function was examined. As demonstrated in Table 3, after accounting for the confounding effects of gender, race and age, smoking by the mother both currently and during pregnancy was significantly associated with a lower FEV,/ FVC. This was not the case for smoking by the father (data not shown), although a dose-response between total household ETS and FEVJFVC was seen. After controlling for the effects of gender, race, height, age and body mass index, maternal smoking during pregnancy was also significantly associated with a lower FEV,. Smoking by the mother both currently and during pregnancy were clearly lower among the most favored SES categories, though the relation of maternal smoking with SES was not monotonic. For SES categories I-V, the percentage of currently smoking mothers were: 52.1%, 34.8%, 43.5%, 33.3% and 28.5%, respectively, P
1. Descriptive characteristics of responders to initial questionnaire
TABLE
Percentage of n=llll Gender Boys Age (years) at testing 5-6 7-11 12-13 Symptoms/diagnosis Usual cough Ever wheeze History of asthma Asthma treatment Parental smoking, current Neither* Mother only* Father only* Both Asthma in a parent Child shares bedroom Child attended day-care Pets in home, current Any Cats
49.8 21.2 76.4 2.4 10.2 10.5 12.0 4.5 48.4 18.3 16.2 17.1 11.3 43.8 33.6 45.3 18.8
*3 18 (28.6%) are monoparental. to be lower with electric baseboard heating (Table 3) the use of which was more common in less favored categories of SES (the percentages in SES categories I to V were: 67.1%, 60.6%, 62.3%, 53.7% and 39.8%, respectively, WO.01). Most subjects had electric stoves for cooking and no effect on lung function of use of different cooking fuels could be demonstrated. There was, however, a significant decreasein airway caliber (FEVJFVC) in the presence of a cat in the home (Table 3). The current presence of cats in the home and day-care attendance were not monotonically related to the SES categories used. The percentage of homes with a cat across SES categories I to V were: 23.1%, 12.8%, 22.3%, 22.6% and 18.7%, respectively. The percentage of children who attended day-care, were: 34.2%, 26.1%, 31.8%, 35.4% and 43.6%, respectively. The estimated percent predicted FEV,, FVC and FEV,/ FVC associated with some domestic factors, day care attendance and asthma in a parent after adjusting for SES and other important confounding variables for both sexes combined and separately for boys and girls are presented in Table 4. In the overall sample (both sexescombined) cat in the home and electric baseboard units remained independently and significantly associated with a lower FEV,/FVC, while day-care attendance significantly reduced FEV, Household ETS was significantly associated with increasing level of FVC. It should be noted that household ETS and maternal smoking during pregnancy were significantly
DOMESTICFACTORSAND TABLE
2. Size measurements
and lung function
levels (crude and standardized)
by demographic
FEV, Demographic variables Age in years 556 7711 12-13 Gender Girls Boys Race Caucasian Non-Caucasian *Standardized tstandardized IStandardized SStandardized TStandardized Standardized
for for for for for for
LUNG FUNCTION
IN CHILDREN
931
variables
FVC
FEVJFVC
Height (cm) mean & SD
BMI mean i SD
1
Percent predicted
1
Percent predicted
Ratio
Percent predicted
120.8 (6.4) 138.9 (9.8) 150.5 (6.6)
15.8 (2.0) 18.9 (3.4) 20.2 (4.1)
1.41 (0.23) 2.03 (0.41) 2.64 (0.42)
96.78* 100.52* 106.45*
1.54 (0.27) 2.55 (0.48) 3.00 (0.49)
96.3* 100.67* 105.49*
0.93 (0.17) 0.90 (0.14) 0.89 (0.15)
95.67t 92.89: 92.26t
134.8 (12.4) 135.7 (10.8)
17.8 (3.4) 17.5 (3.4)
1.88 (0.47) 1.98 (0.45)
100~00~ loo~oog
2.08 (0.54) 2.26 (0.54)
100~00~ 100~00~
0.92 (0.15) 0.89 (0.15)
98.995 96.05
135.1 (11.7) 135.7 (11.6)
17.6 (3.4) 17.9 (3.4)
1.96 (0.46) 1.84 (0.47)
100~00~ 1oo.ofl
2.21 (0.54) 2.04 (0.55)
1oo.oog 1oo.oq
0.90 (0.15) 0.92 (0.17)
96.931, 98.701;
the the the the the the
effects effects effects effects effects effects
of of of of of of
gender, race, height and body mass index. gender and race. age, race, height and body mass index. age and race. age, gender, height and body mass index. age and gender.
correlated k2=240, P
Discussion Exposure to environmental tobacco smoke was found to be associated with a lower FEV,/FVC, only among boys. Several previous studies (3141) have reported reduced lung function with exposure to household ETS, while four cohort (12,4244) and three cross-sectional studies (4547) were unable to demonstrate such an association. Where an effect of ETS has been found, it appeared to be more pronounced among boys (33,39). There were important gender differences in several factors associated with adverse effects on lung size or airway caliber including indoor pollutants. Such gender differences have been reported frequently, especially for ETS, to which boys appear more sensitive than girls. This may be due to differences in maturation rate of the lungs in boys and girls, resulting in differing susceptibilities at various stages of childhood (48).
We cannot rule out the possibility that gender may in some way determine the quality or quantity of various indoor exposures, however. The larger FVC among boys exposed to ETS has also been described previously (23,49). A potential explanation is the inheritance of large lungs from parents who smoke given the evidence that there is health selection into the smoking habit on the basis of larger lung volumes (50). Alternatively, their large FVC may represent a compensatory response to airflow limitation in the maturing lung (51). In contrast to several previous studies (18,52), we failed to find any relationship between type of cooking fuel and lung function. The number of study subjects was smaller than several previous studies, however, and the great majority of our study subjects (more than 90%) used electricity for cooking in their homes. We did find a decrease in airway caliber in relation to electric baseboard heating even after adjusting for SES. Since such heating is more often found in less favored homes in our area, this relationship may be due to persistent confounding by SES due to imprecision in the score used. However, similar findings have been reported previously, for instance, in a recently reported Montreal study which showed an increased incidence of asthma in children living in homes with electric heating (53). Others have found a lower FEV, in association with central heating (54), or hot water heating (18). It is difficult to reconcile these reports; it should be noted that type of heating was not a hypothesized primary risk factor. A potential mechanism has been proposed by Willoughby (55) who suggested that strong electrical current (in this case, produced by thunder storms) might provoke electric charges on allergenic or irritant particles, thus rendering them more reactive or potent.
932
K. DEMISSIE ET AL.
TABLE3. Significance of effect on percent predicted FEV,, FVC, and FEVJFVC of indoor home environmental variables n
FEV,* % (95% CIS)
FVC* % (95% CIS)
FEV,/FVCt % (95% CIS)
Maternal smoking status Never Ex Current
442 74 348
100.32 (99.22-101.44) 98.56 (95.93-101.27) 99.28 (98.05-100.53) P=O.3116
99.35 (98.30--100.41) 97.78 (9528-100.35) 100.72 (9952-101.92) P=O.O706
99.85 (99.25-100.44) 99.71 (98.26101.15) 97.63 (96.97-98.30) P=o~ooo1
Household ETS None (0 cig day - ‘) Light ( I 10 cig day - ‘) Moderate (1 l-24 cig day - ‘) Heavy ( 2 25 cig day - i)
464 146 149 186
100.04 (98.97-101.11) 102.11 (100~19-104~07) 99.28 (97.43-101.16) 98.79 (97.14-100.46) P=O.O661
99.26 (98.25-100.29) 101.31 (99.46-103.19) 100.53 (98.71-102.37) 100.47 (98.85-102.12) P=O.2092
99.64 (99.07-100.22) 99.79 (98.76-100.82) 97.89 (96.87-98.91) 97.33 (96.42-98.24) P=o~ooo1
Smoking during pregnancy Yes No
278 598
98.91 (97.55-100.30) 100.66 (99.71-101.62) P=O.O412
100.07 (98.73-101.42) 99.96 (99.06-100.88) P=O.9012
97.86 (97.1 l-98.60) 99.54 (99~03-100~05) P=O.O003
Type of heating Water radiator Forced air Electric baseboard Others
199 108 545 105
101.48 (99.86-103.12) 100.29 (98.12-102.52) 99.35 (98.38-100.33) 100.75 (98.54-103.01) P=O.1480
100.49 (98.93-102.08) 99.75 (97.64-101.89) 99.87 (98.93-100.82) 100.36 (98.21-102.55) P=O.8963
100.02 (99.13-100.91) 99.50 (98.28-100.71) 98.42 (97.88-98.96) 99.30 (98.07-100.53) P=O.O165
Type of cooking fuel Electricity Others
889 69
99.90 (99.14100.68) 101.50 (98.74-104.34) P=O.2788
99.92 (99.19-100.67) 101.13 (98.46103.86) P=O.3982
98.93 (98.51-99.36) 99.47 (97.95-100.99) P=O.5065
Cat currently in home Present Absent
185 71s
99.24 (97.59-100.93) 100.20 (99.39-101.03) P=O.3141
100.63 (99.00-102.28) 99.85 (99.05-100.65) P=O.3999
97.70 (96.77-98.63) 99.28 (98.83-99.73) P=O.O027
*Standardized for the effects of gender, race, height, age and body mass index. TStandardized for the effects of gender, race and-age. SConfidence interval. A low FEV,IFVC in relation to the presence of a cat in the home may be due to allergic asthma, diagnosed or not, since this potent allergen is a strong risk factor for allergic airway disease (17). This indoor allergen may be especially important in northern climates where dust mite contamination may be relatively low (56). Among girls, but not boys, day-care attendance resulted in a lower FEV, and FEV,/FVC. Others have reported day-care center attendance to be a risk factor for lower respiratory tract illness (57) and this in turn could result in reduced lung function. While we have no explanation for the gender differences in the effect of day-care attendance, a recent study also reported the effect of day-care attendance on lower respiratory illness (LRI) was higher in girls than in boys (58). We used relatively lax criteria for acceptable spirometric tests in this study, in order to include as many children as possible. This is because of an earlier work in our laboratory that demonstrated a strong relationship between respiratory ill health, airway responsiveness to methacholine and spirometric test failure (59,60). To adjust for major
determinants of lung function, such as age and size,we used internal standards as opposed to external standards. The use of an internal standard has been recommended most often (61). We used the residuals from a prediction model for the analysis of lung function data. This model was chosen in order to satisfy the usual modeling assumptions (8,23). In Canada, there is universal accessto health care and the cost of medical services per capita has not been found to vary significantly with income (62). The observed effects of domestic factors on lung function could not be explained on the basis of differential accessto health care, asthma in a parent or socioeconomic status. In conclusion, this study adds further evidence that domestic factors such as exposure to environmental tobacco smoke, the presence of a cat in the home and electric baseboard units for heating are important risk factors for reduced airway caliber. Deficits in lung function are likely to persist into adult life at which time they are associated with an excessin morbidity and mortality (63). Specific corrective actions in order to reduce these exposures may have public health significance.
DOMESTIC TABLE
FACTORS
AND LUNG FUNCTION
IN CHILDREN
933
4. Results of multiple regression analysis with percent predicted FEV,, FVC, and FEV,IFVC as dependent variables
Household ETSI Both sexescombined Boys Girls Maternal smoking during pregnancy Both sexescombined Boys Girls Type of heating11 Both sexescombined Boys Girls Presenceof a cat Both sexescombined Boys Girls Day - care attendance Both sexescombined Boys Girls Asthma in a parent Both sexescombined Boys Girls
Regression coefficient (95%CI*)
FEV,I
FVC$ Regression coefficient (95% cI*)
FEV,/FVC§ Regression coefficient (95% cI*)
+ 1.78 (- 0.59-+4.16) + 1.70 ( - 1.60-+5.00) + 1.08 ( - 24-+4.58)
+2.86 (+0.55-+517)t +3.97 (+0.86-+7.80)t + 1.32 ( - 2.12-+4.88)
- 1.00 (- 2.3ll+0.32) - 2.13 ( - 4.077 - 0.19)? - 0.31 (- 2.10-+1.47)
- 1.81 (- 3.833+0.21) - 2.25 (- 5.17-+0.67) - 1.24 (- 4.10-+1.63)
- 0.71 (- 2.68-+ 1.25) - 1.31 (- 4.07-+ 1.44) - 0.52 ( - 3.38-+2.35)
- 1.05 (- 2,16-+0.07) - 0.96 ( - 2.688+0.76) - 0.61 ( - 2.077+0.85)
- 1.43 (- 3.477+0.61) - 0.81 (- 3.87-+2.26) - 2.39 (- 5.16-+0.37)
- 0.30 ( - 2.299+ 1.69) +0.96 (- 1.93-+3,85) - 1.86 ( - 4.633+0.90)
- 1.26 (- 2.39-- 0.13)? - 1.76 (- 3.56+0.04) - 0.71 (- 2.212-+0.70)
- 0.36 ( - 2.366+ 1.64) - 0.03 ( - 2.999+ 2.94) - 0.78 ( - 3,50-+ 1.95)
+0.84 ( - 1.1l-+2.79) +2.35 ( - 0.45-+5.15) - 0.34 (- 3.06+2.39)
-1.15(-2.26-0.05)t - 2.19 ( - 3.94 - 0.45)t - 0.50 (- 1.89-+0.89)
- 2.05 ( - 3.71- - 0.40)t - 1.10 (- 3.47-+ 1.27) - 2.92 ( - 5.27- - 0.56)?
- 1.45 (- 3,066+0.16) - 1.48 ( - 3.72-+0.76) - 1.49 ( - 3.85-+0.87)
- 0.59 ( - 1.50-+0.33) +0.54 ( - 0.86+ 1.93) - 1.53 ( - 2,73- - 0.33)+
- 1.40 ( - 3.86+ 1.06) - 2.07 (- 5.322+1.18) +0.80 ( - 3.06-+4.66)
- 0.13 (- 2.52-+2.26) - 0.12 (- 3.19-+2.95) +0.81 ( - 3.055+4.68)
- 1.12 (- 2,47-+0.24) - 1.83 (- 3.74+0.08) +0.25 ( - 1.72-+2.22)
*Confidence interval. fP
SAdjusted for the effects of SES, gender, race, height, age, and body mass index in addition to the variables in the table. §Adjusted for the effects of SES, gender, race, and age in addition to the variables in the table. vhe estimate is related to heavy (225 cig day ~ ‘) household ETS in comparison to no (0 cig day - ‘) household ETS exposure. lIThe estimate is related to use of electric baseboard units in comparison to water radiators.
References 1. Chapin FS. Human Activity Patterns in the City. New York: Wiley-Interscience, 1974. 2. Szalai A, ed. The Use of Time: Daily Activities of Urban and Suburban Populations in Twelve Countries. The Hague: Mouton, 1972. 3. Binder R, Mitchell C, Hosein H, Bouhuys A. Importance of the indoor environment in air pollution exposure. Arch Environ Health 1976; 31: 277-279. 4. Ingram JM, Sporik R, Rose G, Honsinger R, Chapman MD, Platts-Mills TAE. Quantitative assessment of exposure to dog (Can fl) and cat (Fe1 dl) allergens: relation to sensitization and asthma among children living in Los Alamos, New Mexico. J Allergy Clin Immunol 1995; 96: 449456. 5. Farrow A, Greenwood R, Preece S, Golding J. Nitrogen dioxide, the oxides of nitrogen, and infants’ health symptoms. ALSPAC Study Team. Avon longitudinal
Study of Pregnancy and Childhood. Arch Environ 1997; 52: 189-194. Spengler JD, Schwab M, McDermott A, Lambert WE, Samet JM. Nitrogen dioxide and respiratory illness in children. Part IV: effects of housing and meteorologic factors on indoor nitrogen dioxide concentrations. Res Rep -Health Effects Institute 1996; 58: l-29 (discussion 3 l-36). Lambert WE, Samet JM, Hunt WC, Skipper BJ, Schwab M, Spengler JD. Nitrogen dioxide and respiratory illness in children. Part II: assessmentof exposure to nitrogen dioxide. Res Rep ~ Health Efects Institute 1993; 58: 33-50 (discussion 51-80). Cunningham J, Dockery DW, Speizer FE. Maternal Smoking during pregnancy as a predictor of lung function in children. Am J Epidemiol 1994; 139: 1139-l 152. Tager IB, Hanrahan JP, Tosteson TD et al. Lung function, pre- and post-natal smoke exposure, and
Health
6.
7.
8.
9.
934
K. DEMISSIE ET AL
wheezing in the first year of life. Am Rev Respir Dis 1993; 147: 811-817. 10. Cunningham J, Dockery DW, Gold DR, Speizer FE. Racial differences in the association between maternal smoking during pregnancy and lung function in children. Am J Respir Crit Care Med 1995; 152: 565-569. 11. Tager IB, Ngo L, Hanrahan JP. Maternal smoking during pregnancy. Am J Respir Crit Care Med 1995; 152: 977-983.
12. Martinez FD, Cline M, Burrows B. Increased incidence of asthma in children of smoking mothers. Pediatrics 1992; 89: 21-26. 13. Hanrahan JP, Tager IB, Segal MR et al. The effect maternal smoking during pregnancy on early infant lung function. Am Rev Respir Dis 1992; 145: 1129-l 135. 14. Chen Y, Li W-X. The effect of passive smoking on children’s pulmonary function in Shanghai. Am J Public Health 1986;76: 515-518. 15. Charlton A. Children and passive smoking: a review. J Fam Pratt 1994; 38: 267-277. 16. Arshad SH. Pets and atopic disorders in infancy. Br J Clin Pratt 1991; 45: 88-89. 17. Sears MR, Herbison BP, Holdaway MD et al. The relative risks of sensitivity to grass pollen, house dust mite and cat dander in the development of childhood asthma. Clin Exp Allergy 1989; 19: 419424. 18. Hosein HR, Cory P, Robertson JMcD. The effect of domestic factors on respiratory symptoms and FEV,. Znt J Epidemiol 1989; 18: 390-396. 19. Dodge R. The effects of indoor pollution on Arizona children. Arch Environ Health 1982; 37: 151-155. 20. Florey CduV, Melia RJW, Chinn S et al. The relation between respiratory illness in primary school children and the use of gas for cooking. III. Nitrogen dioxide, respiratory illness, and lung function. Znt J Epidemiol 1979; 8: 347-353. 21. Hosein HR, Bouhuys A. Possible environmental hazards of gas cooking. Br Med J 1979: 125. 22. Hassselblad V, Humble CG, Graham MG, Anderson HS. Indoor environmental determinants of lung function in children. Am Rev Respir Dis 1981; 123: 479-485. 23. Ware JH, Dockery DW, Spiro III A et al. Passive smoking, gas cooking, and respiratory health of children living in six cities. Am Rev Respir Dis 1984; 129: 366374. 24. Statistic Canada. Census of Canada 1986. Profiles, Census Tracts. Montreal, Part I and II. 25. American Thoracic Society. Standardization of spirometry - 1987 update. Am Rev Respir Dis 1987; 136: 1285-1298. 26. Kanner RE, Schenker MB, Muiioz A et al. Spirometry in children. Methodology for obtaining optimal results for clinical and epidemiologic studies. Am Rev Respir Dis 1983; 127: 720-724. 27. Statistic Canada. Standard Division: Standard Occupational Classtfication (1980). Catalogue 12-565E. 28. Blishen BR, Carroll WK, Moore C. The 1981 socioeconomic index for occupations in Canada. Canada Rev Sot Anth 1987; 24: 465488.
29. Chinn S, Rona RJ. Height and age adjustment for cross sectional studies of lung function in children age 6-l 1 years. Thorax 1992; 47: 7077714. 30. Osterman JW, Armstrong BG, Ledoux E et al. Comparison of French and English version of the American Thoracic Society Respiratory Questionnaire in a bilingual working population. Znt J Epidemiol 1991; 20: 138-143. 31. Chilmonczyk BA, Salmun LM, Megathlin KN et al. Association between exposure to environmental tobacco smoke and exacerbations of asthma in children. N Engl J Med 1993; 328: 1665-1669. 32. Kovesi T, Corey M, Levison H. Passive smoking and lung function in cystic fibrosis. Am Rev Respir Dis 1993; 148: 12661271. 33. Wang X, Wypij D, Gold DR et al. A longitudinal study of the effects of parental smoking on pulmonary function in children 6-18 years. Am J Respir Crit Care Med 1994; 149: 1420-1425. 34. Casale R, Natali G, Colantonio D, Pasqualetti P. Circadian rhythm of peak expiratory flow in children passively exposed and not exposed to cigarette smoke. Thorax 1992; 47: 801-803. 35. Cook DG, Whincup PH, Papacosta 0, Strachan DP, Jarvid MJ, Bryant A. Relation of passive smoking as assessedby salivary cotinine concentration and questionnaire to spirometric indices in children. Thorax 1993; 48: 14-20. 36. Cunningham J, Dockery DW, Speizer FE. Maternal Smoking during pregnancy as a predictor of lung function in children. Am J Epidemiol 1994; 139: 1139-1152. 37. Forastiere F, Agabiti N, Corbo GM et al. Passive smoking as a determinant of bronchial responsiveness in children. Am J Respir Crit Care Med 1994; 149: 365-370. 38. Frischer T, Kuhr J, Meinek R et al. Influence of maternal smoking on variability of peak expiratory flow rate in school children. Chest 1993; 104: 1133-l 137. 39. Rona RJ, Chinn S. Lung function, respiratory illness, and passive smoking in British primary school children. Thorax 1993; 48: 21-25. 40. Schmitzberger R, Rhomberg K, Buchele H et al. Effects of air pollution on the respiratory tract of children. Pediatr Pulmonol 1993; 15: 68-74. 41. Smyth A, O’Hea U, Williams G, Smyth R, Heaf D. Passive smoking and impaired lung function in cystic fibrosis. Arch Dis Child 1994; 71: 353-354. 42. Kitchen WH, Olinsky A, Doyle LW et al. Respiratory health and lung function in g-year old children of very low birthweight: a cohort study. Pediatrics 1992; 89: 1151-1158. 43. Lebowitz MD, Sherrill D, Holberg CJ. Effects of passive smoking on lung growth in children. Pediatr Pulmonol 1992; 12: 3742. 44. Sherrill DL, Martinez FD, Lebowitz MD et al. Longitudinal effects of passive smoking on pulmonary function in New Zealand children. Am Rev Respir Dis 1992; 145: 1136-1141.
DOMESTIC
45. Goren AI, Hellman S. Passivesmoking among school children in Israel. Env Health Perspect 1991; 96: 203-2 11. 46. Guneser S, Atici A, Alparslan N, Cinaz P. Effects of indoor environmental factors on respiratory systemsof children. J Tropical Pediatr 1994; 40: 114-l 17. 47. Willers S, Attewell R, Bensryd I, Schutz A, Skarping G, Vahter M. Exposure to environmental tobacco smoke in the household and urinary cotinine excretion, heavy metals retention, and lung function. Arch Environ Health
1992; 41: 357-363.
48. Hibbert ME, Courlej JM, Landau Li. Changes in lung, airway and chest wall function in boys and girls between 8 and 12 years. J Appl Physiol 1984; 57: 304-308. 49. Vedal S, Schenker MB, Samet JM et al. Risk factors for childhood respiratory disease: Analysis of pulmonary function. Am Rev Respir Dis 1984; 130: 187-192. 50. Becklake MR, Lalloo UG. The ‘healthy smoker’: a phenomenon of health selection? Respiration 1990; 57: 137-144. 51. Weiss ST, Tosteson TD, Segal MR et al. Effects of asthma on pulmonary function in children. A longitudinal population based study. Am Rev Respir Dis 1992; 1456: 58-64. 52. Melia RJW, Florey C du V, Chinn S. Relation between respiratory illness in primary school children and the use of gas for cooking. Results from a National study. Int J Epidemiol 1979; 8: 3333338. 53. Infante-Rivard C. Childhood asthma and indoor environmental risk factors. Am J Epidemiol 1993; 137: 834844.
FACTORSAND
LUNGFUNCTIONINCHILDREN
935
54. Yarnell JWG, St. Leger AS. Housing conditions, respiratory illness and lung function in children in South Wales. Br J Prev Sot Med 1977; 31: 1833188. 55. Willoughby DA. Asthma increase, summer 1994 (letter to the editor). Lancet 1994; 344: 413. 56. Colloff MJ, Ayres J, Carswell F et al. The control of allergens of dust mites and domestic pets: a position paper. Clin Exp Allergy 1992; 22: l-28. 57. Anderson LJ, Parker RA, Strikas RA et al. Day-care center attendance and hospitalization for lower respiratory tract illness. Pediatrics 1988; 82: 300-308. 58. Marbury MC, Maldonado G, Waller L. Lower respiratory illness, recurrent wheezing, and day care attendance. Am J Respir Crit Care Med 1997; 155: 156-161. 59. Becklake MR. Epidemiology of test failure. Br J Znd Med 1990; 47: 73374. 60. Ng’Ang’a LW, Ernst P, Jaakkola MS, Gerardi G, Hanley JH, Becklake MR. Spirometric lung function. Distribution and determinants of test failure in a young adult population. Am Rev Respir Dis 1992; 145: 48-52. 61. Vollmer WM, Johnson LR, McCamant LE, Buist AS. Methodologic issues in the analysis of lung function data. J Chron Dis 1987; 40: 1013-1023. 62. Manga P. Income and access to medical care in Canada. In: Coburn D, D’Arcy C, New P et al., eds. Health and Canadian Society. Ontario: Fitzhenry & Whiteside, Markham, 1981: 325-342. 63. Barker DJP, Godfrey KM, Fall C et al. Relation of birthweight and childhood respiratory infection to adult lung function and death from chronic obstructive airways disease. Br Med J 1992; 303: 671-675.
MEDICINE
(1998)
92,
928-935
The role of domestic factors and day-care attendance on lung function of primary school children K. DEMISSIE*, P. ERNST*, L. JOSEPH+ AND M. R. BECKLAKE* “Department of Epidemiology and Biostatistics, Respiratory Epidemiology Unit, McGill University, 1110 Pine Avenue West, Montreal, Quebec, Canada H3A 1A3 ‘Departmenf of Epidemiology and Biostaitistics,McGill University, 1020 Pine Avenue West, Monfreal, Quebec, Canada H3A IA2 The results of studies examining the relationship of domestic factors to lung function are contradictory. We therefore examined the independent effects of maternal smoking during pregnancy, exposure to environmental tobacco smoke (ETS), the presence of a cat, type of heating and cooking used in the home and day-care attendance on lung function after controlling for socioeconomic status (SES). Nine hundred and eighty-nine children from 18 Montreal schools were studied between April 1990 and November 1992. Information on the child’s health and exposure to domestic factors was collected by questionnaire. Spirometry was performed at school. The data were analysed by multiple linear regression with percent predicted FEV,, FVC, and FEV,/FVC as dependent variables. In the overall sample (both sexescombined), cat in the home (regression coefficient, p= - 1.15, 95% confidence interval, CI: - 2.26- 0.05) and electric baseboard units (J= - 1.26, 95% CI: - 2.39- - 0.13) were independently associated with a lower FEVJFVC, while day-care attendance (j?= - 2.05, 95% CI: - 3.71- - 0.40) significantly reduced FEV,. Household ETS was significantly associated with increasing level of FVC (p=2.86,95% CI: +0.55 to +5.17). In boys but not girls, household ETS @= - 2.13, 95% CI: - 4.07- - 0.19) and the presence of a cat u= - 2.19, 95% CI: - 3.94- - 0.45) were associated with lower FEV,/FVC. By contrast, day-care attendance was associated with lower FEV, (p= - 2.92, 95% CI: - 5.27- - 0.56) and FEV,/FVC @= - 1.53, 95% CI: - 2.73- - 0.33) in girls only. In conclusion, the results provide evidence that domestic factors and day-care attendance primarily affected airway caliber and gender differences were apparent in the effects of these factors. RESPIR. MED.
(1998)
92,
928-935
Introduction Various indoor and outdoor pollutants have been found to be associated with lower respiratory diseasesof infants and children. Children in Western countries spend more than 80% of their time indoors (1,2). Furthermore, the weight of evidence suggests indoor (as opposed to outdoor) pollutants to be more important as a risk factor for respiratory ill health in children (3). Concentrations of respirable particulate, nitrogen dioxide, and dust allergens are higher in homes where gas is used for cooking and in homes that contain pets (4-7). Received 21 January 1997, and accepted in revised form 26 February 1998. This work was supported by the Respiratory Health Network of Centers of Excellence and the Medical Research Council of Canada. Correspondence should be addressed to: K. Demissie, Department of Family Medicine, UMDNJ-Robert Wood Johnson Medical School, One Robert Wood Johnson Place-CN 19, New Brunswick, NJ 08903-0019, U.S.A. 0954-6111/98/070928+08
$12.0010
Several investigators have found an association between environmental tobacco smoke (ETS) and reduced lung function in the early postnatal period (8-15). Most of these studies were unable to assessthe independent contributions of prenatal and early postnatal exposure to tobacco smoke on lung function. Furthermore, information as to whether a deficit in lrtero or in the early postnatal period persists to childhood or is compensated by a catch-up growth has only been reported infrequently (8). Although the relationships of pet exposure and the development of sensitization and allergic diseaseshave been extensively investigated (16,17), data on the relationship of exposure to pets and lung function is limited. In a study among 1357 non-smoking children in the U.S.A., the presence of indoor pets was not found to be associated with a lower FEV, (18). The relationship between type of home heating and cooking fuel used in the home with children’s lung function levels is less consistent. Several studies of gas cooking and pulmonary function levels have found no association (1921), while Hasselblad et al. (22) and Ware et al. (23) found 0
1998
W. B. SAUNDERS
COMPANY
LTD
DOMESTICFACTORSAND
reduced FEV, among children living in homes with gas stoves. More recently, Hosein et al. (18) reported the use of gas stoves, the absence of air conditioning and the use of hot water heating to be associated with lower FEV,. The type of home heating, use of air conditioning, cooking fuel use and exposure to ETS are likely to differ according to socioeconomic status (SES). Most studies reporting on the effect of such indoor factors did not control for SES or used imprecise methods of assessing social class that are likely to have resulted in residual confounding. We therefore examined the independent effects of maternal smoking during pregnancy, postnatal exposure to ETS, presence of a cat in the home, type of home heating and cooking fuel used in the home, parental asthma, and day-care attendance on lung function in a study of school children after accounting for the possible effects of SES, gender, age, race, height and body mass index of the child.
Methods
LUNG FUNCTION
IN CHILDREN
929
Symptoms and diagnoses were defined as follows: ‘Does the child’s chest ever sound wheezy or whistling? (ever wheeze); ‘Does the child usually cough?’ (usual cough); ‘Has the child ever been diagnosed as having asthma?’ (history of asthma); ‘Does your child take any medications for breathing?’ (asthma treatment); ‘Does the child’s natural father suffer from asthma?‘/‘Does the child’s natural mother suffer from asthma?’ (asthma in a parent). LUNG FUNCTION In the school gymnasium, each subject’s age, gender, height, and weight were recorded, and the children were asked whether they had experienced a respiratory tract infection within the previous week. Spirometry was carried out sitting and with nose clips using two Collins 10-l water-sealed spirometers (Warren E. Collins, Braintree, MA, U.S.A.) according to current American Thoracic Society guidelines (25). The best FEV, and FVC from any acceptable flow-volume curve were used for analysis (26).
STUDY POPULATION
ASSESSMENT OF SOCIOECONOMIC
Eighteen schools were selected on the island of Montreal in order to represent a broad range of socioeconomic status (SES).To achieve this, all schools in the five school boards in central Montreal were ranked according to neighborhood average house values published by Statistics Canada (24). Within each school board, schools were selected from the upper, middle and lower ranges of neighborhood average house value. For each of the 18 schools so selected, three classesfrom each of grades one (5-7 years of age), three (8 and 9 years of age) and five (lo-13 years of age) were selected. If more than one class of a grade was available, we took the first in which the teacher agreed to participate and approached the teachers at their convenience or in the order suggested by the school principal. A total of 1274 children were recruited for the study. Comparison of schools selected (n=18) with those not selected (n=291) showed no meaningful difference with respect to neighborhood poverty, income and educational attainment (24).
Parental occupation was transformed into an SES score by coding the last occupation of the parents into the corresponding codes of the Canadian Classification and Dictionary of Occupations (27). These codes were then converted into SES scores (the highest score from either parent was retained for analysis) for the child, based on the income and education level for each occupation from the tables developed by Blishen et al. (28). The SES score was used to group the children into five categories. Children from unemployed families were grouped into category I, whereas categories II to V were formed using the quartiles of the SES score, category II with the lowest and category V with the highest score. respectively.
RESPIRATORY QUESTIONNAIRE A questionnaire about the child’s health and home environment was completed by parents (in 73% by biological mothers, 24% by biological fathers, and 3% by either a step or grandparent). Parents were asked about the current presence of pets (dogs, cats, birds, hamsters/mice/guinea pigs) in the home, the type of cooking and heating fuel used in the home, the number of people sleeping in the child’s bedroom (child shares bedroom) and whether the child had attended day-care. The child’s exposure to environmental tobacco smoke both in utero (maternal smoking during pregnancy) and postnatally was also assessed, and an attempt was made to quantify current smoking exposure by determining the number of smokers in the home and their average number of cigarettes smoked per day at home (household ETS).
STATISTICAL
STATUS
METHODS
The outcome variables were FEV,, FVC, and the ratio of FEV, to FVC (FEV,/FVC). The explanatory variables considered in this analysis were maternal smoking during pregnancy, household ETS, type of home heating and cooking fuel, current presence of pets, asthma in a parent, day-care attendance, gender, age, race, height and body mass index of the child. In children, it has been recommended not to use multiple regression on untransformed lung function data (29). To standardize the lung function values, FEV, and FVC were each transformed into a natural logarithmic scale (In) and regressed on In (standing height), In (age), and In (body mass index), gender and race. Constant variance and normality was obtained in the residuals. The antilogarithm of the residuals multiplied by 100 provided the percent predicted lung function values (using an internal standard value obtained among all children). The relationships of the percent predicted lung function values to the categorical independent variables were evaluated by multiple linear regression before and after accounting for the effects of
930
K. DEMISSIE ET AL.
potentially confounding variables. For FEV,/FVC, normality and homogeneity of variance could be obtained by the arc sine transformation. The equation for the calculation of percent predicted FEV,/FVC was obtained from this transformed variable. Multiple linear regression was used to evaluate the relationship between the independent variables and percent predicted FEV,/FVC after back transforming the ratio to the original scale.
Results A total of 1274 children were eligible to participate in the study from the 18 selected Montreal schools. Of this, the parents of 130 (10.2%) refused participation for their children, whereas a further 75 (59%) children did not return the questionnaire and consent form. Spirometry was not performed for a further 23 children because they were absent from school. Of the 1046 children who performed the spirometric test, 28 (2.7%) did not have acceptable spirometric data. The spirometry data of 28 (3%) children were lost after the test. One child (0.1%) was excluded because of severe asthmatic attack at the time of the test. There were no meaningful differences between participants and non participants as to the age of the child, sex and socioeconomic status as assessedby neighborhood census data. After the above exclusions, 989 (78%) children remained for analysis. Seven hundred and sixty-six (78%) of these children were Caucasian and ranged in age from 5 to 13 years, though the great majority (99.4%) were 6 to 12 years of age. Descriptive characteristics of the children whose parents responded to the initial questionnaire are presented in Table 1. The prevalence of wheezing was quite low in comparison to published results in similar populations. This is likely to be due to the unsatisfactory translation of the term wheezing into French and has been described previously in studies carried out in the province of Quebec (30). Size measurements and lung function levels (crude and standardized) as well as demographic variables are displayed in Table 2. The effect of different exposures in the home on lung function was examined. As demonstrated in Table 3, after accounting for the confounding effects of gender, race and age, smoking by the mother both currently and during pregnancy was significantly associated with a lower FEV,/ FVC. This was not the case for smoking by the father (data not shown), although a dose-response between total household ETS and FEVJFVC was seen. After controlling for the effects of gender, race, height, age and body mass index, maternal smoking during pregnancy was also significantly associated with a lower FEV,. Smoking by the mother both currently and during pregnancy were clearly lower among the most favored SES categories, though the relation of maternal smoking with SES was not monotonic. For SES categories I-V, the percentage of currently smoking mothers were: 52.1%, 34.8%, 43.5%, 33.3% and 28.5%, respectively, P
1. Descriptive characteristics of responders to initial questionnaire
TABLE
Percentage of n=llll Gender Boys Age (years) at testing 5-6 7-11 12-13 Symptoms/diagnosis Usual cough Ever wheeze History of asthma Asthma treatment Parental smoking, current Neither* Mother only* Father only* Both Asthma in a parent Child shares bedroom Child attended day-care Pets in home, current Any Cats
49.8 21.2 76.4 2.4 10.2 10.5 12.0 4.5 48.4 18.3 16.2 17.1 11.3 43.8 33.6 45.3 18.8
*3 18 (28.6%) are monoparental. to be lower with electric baseboard heating (Table 3) the use of which was more common in less favored categories of SES (the percentages in SES categories I to V were: 67.1%, 60.6%, 62.3%, 53.7% and 39.8%, respectively, WO.01). Most subjects had electric stoves for cooking and no effect on lung function of use of different cooking fuels could be demonstrated. There was, however, a significant decreasein airway caliber (FEVJFVC) in the presence of a cat in the home (Table 3). The current presence of cats in the home and day-care attendance were not monotonically related to the SES categories used. The percentage of homes with a cat across SES categories I to V were: 23.1%, 12.8%, 22.3%, 22.6% and 18.7%, respectively. The percentage of children who attended day-care, were: 34.2%, 26.1%, 31.8%, 35.4% and 43.6%, respectively. The estimated percent predicted FEV,, FVC and FEV,/ FVC associated with some domestic factors, day care attendance and asthma in a parent after adjusting for SES and other important confounding variables for both sexes combined and separately for boys and girls are presented in Table 4. In the overall sample (both sexescombined) cat in the home and electric baseboard units remained independently and significantly associated with a lower FEV,/FVC, while day-care attendance significantly reduced FEV, Household ETS was significantly associated with increasing level of FVC. It should be noted that household ETS and maternal smoking during pregnancy were significantly
DOMESTICFACTORSAND TABLE
2. Size measurements
and lung function
levels (crude and standardized)
by demographic
FEV, Demographic variables Age in years 556 7711 12-13 Gender Girls Boys Race Caucasian Non-Caucasian *Standardized tstandardized IStandardized SStandardized TStandardized Standardized
for for for for for for
LUNG FUNCTION
IN CHILDREN
931
variables
FVC
FEVJFVC
Height (cm) mean & SD
BMI mean i SD
1
Percent predicted
1
Percent predicted
Ratio
Percent predicted
120.8 (6.4) 138.9 (9.8) 150.5 (6.6)
15.8 (2.0) 18.9 (3.4) 20.2 (4.1)
1.41 (0.23) 2.03 (0.41) 2.64 (0.42)
96.78* 100.52* 106.45*
1.54 (0.27) 2.55 (0.48) 3.00 (0.49)
96.3* 100.67* 105.49*
0.93 (0.17) 0.90 (0.14) 0.89 (0.15)
95.67t 92.89: 92.26t
134.8 (12.4) 135.7 (10.8)
17.8 (3.4) 17.5 (3.4)
1.88 (0.47) 1.98 (0.45)
100~00~ loo~oog
2.08 (0.54) 2.26 (0.54)
100~00~ 100~00~
0.92 (0.15) 0.89 (0.15)
98.995 96.05
135.1 (11.7) 135.7 (11.6)
17.6 (3.4) 17.9 (3.4)
1.96 (0.46) 1.84 (0.47)
100~00~ 1oo.ofl
2.21 (0.54) 2.04 (0.55)
1oo.oog 1oo.oq
0.90 (0.15) 0.92 (0.17)
96.931, 98.701;
the the the the the the
effects effects effects effects effects effects
of of of of of of
gender, race, height and body mass index. gender and race. age, race, height and body mass index. age and race. age, gender, height and body mass index. age and gender.
correlated k2=240, P
Discussion Exposure to environmental tobacco smoke was found to be associated with a lower FEV,/FVC, only among boys. Several previous studies (3141) have reported reduced lung function with exposure to household ETS, while four cohort (12,4244) and three cross-sectional studies (4547) were unable to demonstrate such an association. Where an effect of ETS has been found, it appeared to be more pronounced among boys (33,39). There were important gender differences in several factors associated with adverse effects on lung size or airway caliber including indoor pollutants. Such gender differences have been reported frequently, especially for ETS, to which boys appear more sensitive than girls. This may be due to differences in maturation rate of the lungs in boys and girls, resulting in differing susceptibilities at various stages of childhood (48).
We cannot rule out the possibility that gender may in some way determine the quality or quantity of various indoor exposures, however. The larger FVC among boys exposed to ETS has also been described previously (23,49). A potential explanation is the inheritance of large lungs from parents who smoke given the evidence that there is health selection into the smoking habit on the basis of larger lung volumes (50). Alternatively, their large FVC may represent a compensatory response to airflow limitation in the maturing lung (51). In contrast to several previous studies (18,52), we failed to find any relationship between type of cooking fuel and lung function. The number of study subjects was smaller than several previous studies, however, and the great majority of our study subjects (more than 90%) used electricity for cooking in their homes. We did find a decrease in airway caliber in relation to electric baseboard heating even after adjusting for SES. Since such heating is more often found in less favored homes in our area, this relationship may be due to persistent confounding by SES due to imprecision in the score used. However, similar findings have been reported previously, for instance, in a recently reported Montreal study which showed an increased incidence of asthma in children living in homes with electric heating (53). Others have found a lower FEV, in association with central heating (54), or hot water heating (18). It is difficult to reconcile these reports; it should be noted that type of heating was not a hypothesized primary risk factor. A potential mechanism has been proposed by Willoughby (55) who suggested that strong electrical current (in this case, produced by thunder storms) might provoke electric charges on allergenic or irritant particles, thus rendering them more reactive or potent.
932
K. DEMISSIE ET AL.
TABLE3. Significance of effect on percent predicted FEV,, FVC, and FEVJFVC of indoor home environmental variables n
FEV,* % (95% CIS)
FVC* % (95% CIS)
FEV,/FVCt % (95% CIS)
Maternal smoking status Never Ex Current
442 74 348
100.32 (99.22-101.44) 98.56 (95.93-101.27) 99.28 (98.05-100.53) P=O.3116
99.35 (98.30--100.41) 97.78 (9528-100.35) 100.72 (9952-101.92) P=O.O706
99.85 (99.25-100.44) 99.71 (98.26101.15) 97.63 (96.97-98.30) P=o~ooo1
Household ETS None (0 cig day - ‘) Light ( I 10 cig day - ‘) Moderate (1 l-24 cig day - ‘) Heavy ( 2 25 cig day - i)
464 146 149 186
100.04 (98.97-101.11) 102.11 (100~19-104~07) 99.28 (97.43-101.16) 98.79 (97.14-100.46) P=O.O661
99.26 (98.25-100.29) 101.31 (99.46-103.19) 100.53 (98.71-102.37) 100.47 (98.85-102.12) P=O.2092
99.64 (99.07-100.22) 99.79 (98.76-100.82) 97.89 (96.87-98.91) 97.33 (96.42-98.24) P=o~ooo1
Smoking during pregnancy Yes No
278 598
98.91 (97.55-100.30) 100.66 (99.71-101.62) P=O.O412
100.07 (98.73-101.42) 99.96 (99.06-100.88) P=O.9012
97.86 (97.1 l-98.60) 99.54 (99~03-100~05) P=O.O003
Type of heating Water radiator Forced air Electric baseboard Others
199 108 545 105
101.48 (99.86-103.12) 100.29 (98.12-102.52) 99.35 (98.38-100.33) 100.75 (98.54-103.01) P=O.1480
100.49 (98.93-102.08) 99.75 (97.64-101.89) 99.87 (98.93-100.82) 100.36 (98.21-102.55) P=O.8963
100.02 (99.13-100.91) 99.50 (98.28-100.71) 98.42 (97.88-98.96) 99.30 (98.07-100.53) P=O.O165
Type of cooking fuel Electricity Others
889 69
99.90 (99.14100.68) 101.50 (98.74-104.34) P=O.2788
99.92 (99.19-100.67) 101.13 (98.46103.86) P=O.3982
98.93 (98.51-99.36) 99.47 (97.95-100.99) P=O.5065
Cat currently in home Present Absent
185 71s
99.24 (97.59-100.93) 100.20 (99.39-101.03) P=O.3141
100.63 (99.00-102.28) 99.85 (99.05-100.65) P=O.3999
97.70 (96.77-98.63) 99.28 (98.83-99.73) P=O.O027
*Standardized for the effects of gender, race, height, age and body mass index. TStandardized for the effects of gender, race and-age. SConfidence interval. A low FEV,IFVC in relation to the presence of a cat in the home may be due to allergic asthma, diagnosed or not, since this potent allergen is a strong risk factor for allergic airway disease (17). This indoor allergen may be especially important in northern climates where dust mite contamination may be relatively low (56). Among girls, but not boys, day-care attendance resulted in a lower FEV, and FEV,/FVC. Others have reported day-care center attendance to be a risk factor for lower respiratory tract illness (57) and this in turn could result in reduced lung function. While we have no explanation for the gender differences in the effect of day-care attendance, a recent study also reported the effect of day-care attendance on lower respiratory illness (LRI) was higher in girls than in boys (58). We used relatively lax criteria for acceptable spirometric tests in this study, in order to include as many children as possible. This is because of an earlier work in our laboratory that demonstrated a strong relationship between respiratory ill health, airway responsiveness to methacholine and spirometric test failure (59,60). To adjust for major
determinants of lung function, such as age and size,we used internal standards as opposed to external standards. The use of an internal standard has been recommended most often (61). We used the residuals from a prediction model for the analysis of lung function data. This model was chosen in order to satisfy the usual modeling assumptions (8,23). In Canada, there is universal accessto health care and the cost of medical services per capita has not been found to vary significantly with income (62). The observed effects of domestic factors on lung function could not be explained on the basis of differential accessto health care, asthma in a parent or socioeconomic status. In conclusion, this study adds further evidence that domestic factors such as exposure to environmental tobacco smoke, the presence of a cat in the home and electric baseboard units for heating are important risk factors for reduced airway caliber. Deficits in lung function are likely to persist into adult life at which time they are associated with an excessin morbidity and mortality (63). Specific corrective actions in order to reduce these exposures may have public health significance.
DOMESTIC TABLE
FACTORS
AND LUNG FUNCTION
IN CHILDREN
933
4. Results of multiple regression analysis with percent predicted FEV,, FVC, and FEV,IFVC as dependent variables
Household ETSI Both sexescombined Boys Girls Maternal smoking during pregnancy Both sexescombined Boys Girls Type of heating11 Both sexescombined Boys Girls Presenceof a cat Both sexescombined Boys Girls Day - care attendance Both sexescombined Boys Girls Asthma in a parent Both sexescombined Boys Girls
Regression coefficient (95%CI*)
FEV,I
FVC$ Regression coefficient (95% cI*)
FEV,/FVC§ Regression coefficient (95% cI*)
+ 1.78 (- 0.59-+4.16) + 1.70 ( - 1.60-+5.00) + 1.08 ( - 24-+4.58)
+2.86 (+0.55-+517)t +3.97 (+0.86-+7.80)t + 1.32 ( - 2.12-+4.88)
- 1.00 (- 2.3ll+0.32) - 2.13 ( - 4.077 - 0.19)? - 0.31 (- 2.10-+1.47)
- 1.81 (- 3.833+0.21) - 2.25 (- 5.17-+0.67) - 1.24 (- 4.10-+1.63)
- 0.71 (- 2.68-+ 1.25) - 1.31 (- 4.07-+ 1.44) - 0.52 ( - 3.38-+2.35)
- 1.05 (- 2,16-+0.07) - 0.96 ( - 2.688+0.76) - 0.61 ( - 2.077+0.85)
- 1.43 (- 3.477+0.61) - 0.81 (- 3.87-+2.26) - 2.39 (- 5.16-+0.37)
- 0.30 ( - 2.299+ 1.69) +0.96 (- 1.93-+3,85) - 1.86 ( - 4.633+0.90)
- 1.26 (- 2.39-- 0.13)? - 1.76 (- 3.56+0.04) - 0.71 (- 2.212-+0.70)
- 0.36 ( - 2.366+ 1.64) - 0.03 ( - 2.999+ 2.94) - 0.78 ( - 3,50-+ 1.95)
+0.84 ( - 1.1l-+2.79) +2.35 ( - 0.45-+5.15) - 0.34 (- 3.06+2.39)
-1.15(-2.26-0.05)t - 2.19 ( - 3.94 - 0.45)t - 0.50 (- 1.89-+0.89)
- 2.05 ( - 3.71- - 0.40)t - 1.10 (- 3.47-+ 1.27) - 2.92 ( - 5.27- - 0.56)?
- 1.45 (- 3,066+0.16) - 1.48 ( - 3.72-+0.76) - 1.49 ( - 3.85-+0.87)
- 0.59 ( - 1.50-+0.33) +0.54 ( - 0.86+ 1.93) - 1.53 ( - 2,73- - 0.33)+
- 1.40 ( - 3.86+ 1.06) - 2.07 (- 5.322+1.18) +0.80 ( - 3.06-+4.66)
- 0.13 (- 2.52-+2.26) - 0.12 (- 3.19-+2.95) +0.81 ( - 3.055+4.68)
- 1.12 (- 2,47-+0.24) - 1.83 (- 3.74+0.08) +0.25 ( - 1.72-+2.22)
*Confidence interval. fP
SAdjusted for the effects of SES, gender, race, height, age, and body mass index in addition to the variables in the table. §Adjusted for the effects of SES, gender, race, and age in addition to the variables in the table. vhe estimate is related to heavy (225 cig day ~ ‘) household ETS in comparison to no (0 cig day - ‘) household ETS exposure. lIThe estimate is related to use of electric baseboard units in comparison to water radiators.
References 1. Chapin FS. Human Activity Patterns in the City. New York: Wiley-Interscience, 1974. 2. Szalai A, ed. The Use of Time: Daily Activities of Urban and Suburban Populations in Twelve Countries. The Hague: Mouton, 1972. 3. Binder R, Mitchell C, Hosein H, Bouhuys A. Importance of the indoor environment in air pollution exposure. Arch Environ Health 1976; 31: 277-279. 4. Ingram JM, Sporik R, Rose G, Honsinger R, Chapman MD, Platts-Mills TAE. Quantitative assessment of exposure to dog (Can fl) and cat (Fe1 dl) allergens: relation to sensitization and asthma among children living in Los Alamos, New Mexico. J Allergy Clin Immunol 1995; 96: 449456. 5. Farrow A, Greenwood R, Preece S, Golding J. Nitrogen dioxide, the oxides of nitrogen, and infants’ health symptoms. ALSPAC Study Team. Avon longitudinal
Study of Pregnancy and Childhood. Arch Environ 1997; 52: 189-194. Spengler JD, Schwab M, McDermott A, Lambert WE, Samet JM. Nitrogen dioxide and respiratory illness in children. Part IV: effects of housing and meteorologic factors on indoor nitrogen dioxide concentrations. Res Rep -Health Effects Institute 1996; 58: l-29 (discussion 3 l-36). Lambert WE, Samet JM, Hunt WC, Skipper BJ, Schwab M, Spengler JD. Nitrogen dioxide and respiratory illness in children. Part II: assessmentof exposure to nitrogen dioxide. Res Rep ~ Health Efects Institute 1993; 58: 33-50 (discussion 51-80). Cunningham J, Dockery DW, Speizer FE. Maternal Smoking during pregnancy as a predictor of lung function in children. Am J Epidemiol 1994; 139: 1139-l 152. Tager IB, Hanrahan JP, Tosteson TD et al. Lung function, pre- and post-natal smoke exposure, and
Health
6.
7.
8.
9.
934
K. DEMISSIE ET AL
wheezing in the first year of life. Am Rev Respir Dis 1993; 147: 811-817. 10. Cunningham J, Dockery DW, Gold DR, Speizer FE. Racial differences in the association between maternal smoking during pregnancy and lung function in children. Am J Respir Crit Care Med 1995; 152: 565-569. 11. Tager IB, Ngo L, Hanrahan JP. Maternal smoking during pregnancy. Am J Respir Crit Care Med 1995; 152: 977-983.
12. Martinez FD, Cline M, Burrows B. Increased incidence of asthma in children of smoking mothers. Pediatrics 1992; 89: 21-26. 13. Hanrahan JP, Tager IB, Segal MR et al. The effect maternal smoking during pregnancy on early infant lung function. Am Rev Respir Dis 1992; 145: 1129-l 135. 14. Chen Y, Li W-X. The effect of passive smoking on children’s pulmonary function in Shanghai. Am J Public Health 1986;76: 515-518. 15. Charlton A. Children and passive smoking: a review. J Fam Pratt 1994; 38: 267-277. 16. Arshad SH. Pets and atopic disorders in infancy. Br J Clin Pratt 1991; 45: 88-89. 17. Sears MR, Herbison BP, Holdaway MD et al. The relative risks of sensitivity to grass pollen, house dust mite and cat dander in the development of childhood asthma. Clin Exp Allergy 1989; 19: 419424. 18. Hosein HR, Cory P, Robertson JMcD. The effect of domestic factors on respiratory symptoms and FEV,. Znt J Epidemiol 1989; 18: 390-396. 19. Dodge R. The effects of indoor pollution on Arizona children. Arch Environ Health 1982; 37: 151-155. 20. Florey CduV, Melia RJW, Chinn S et al. The relation between respiratory illness in primary school children and the use of gas for cooking. III. Nitrogen dioxide, respiratory illness, and lung function. Znt J Epidemiol 1979; 8: 347-353. 21. Hosein HR, Bouhuys A. Possible environmental hazards of gas cooking. Br Med J 1979: 125. 22. Hassselblad V, Humble CG, Graham MG, Anderson HS. Indoor environmental determinants of lung function in children. Am Rev Respir Dis 1981; 123: 479-485. 23. Ware JH, Dockery DW, Spiro III A et al. Passive smoking, gas cooking, and respiratory health of children living in six cities. Am Rev Respir Dis 1984; 129: 366374. 24. Statistic Canada. Census of Canada 1986. Profiles, Census Tracts. Montreal, Part I and II. 25. American Thoracic Society. Standardization of spirometry - 1987 update. Am Rev Respir Dis 1987; 136: 1285-1298. 26. Kanner RE, Schenker MB, Muiioz A et al. Spirometry in children. Methodology for obtaining optimal results for clinical and epidemiologic studies. Am Rev Respir Dis 1983; 127: 720-724. 27. Statistic Canada. Standard Division: Standard Occupational Classtfication (1980). Catalogue 12-565E. 28. Blishen BR, Carroll WK, Moore C. The 1981 socioeconomic index for occupations in Canada. Canada Rev Sot Anth 1987; 24: 465488.
29. Chinn S, Rona RJ. Height and age adjustment for cross sectional studies of lung function in children age 6-l 1 years. Thorax 1992; 47: 7077714. 30. Osterman JW, Armstrong BG, Ledoux E et al. Comparison of French and English version of the American Thoracic Society Respiratory Questionnaire in a bilingual working population. Znt J Epidemiol 1991; 20: 138-143. 31. Chilmonczyk BA, Salmun LM, Megathlin KN et al. Association between exposure to environmental tobacco smoke and exacerbations of asthma in children. N Engl J Med 1993; 328: 1665-1669. 32. Kovesi T, Corey M, Levison H. Passive smoking and lung function in cystic fibrosis. Am Rev Respir Dis 1993; 148: 12661271. 33. Wang X, Wypij D, Gold DR et al. A longitudinal study of the effects of parental smoking on pulmonary function in children 6-18 years. Am J Respir Crit Care Med 1994; 149: 1420-1425. 34. Casale R, Natali G, Colantonio D, Pasqualetti P. Circadian rhythm of peak expiratory flow in children passively exposed and not exposed to cigarette smoke. Thorax 1992; 47: 801-803. 35. Cook DG, Whincup PH, Papacosta 0, Strachan DP, Jarvid MJ, Bryant A. Relation of passive smoking as assessedby salivary cotinine concentration and questionnaire to spirometric indices in children. Thorax 1993; 48: 14-20. 36. Cunningham J, Dockery DW, Speizer FE. Maternal Smoking during pregnancy as a predictor of lung function in children. Am J Epidemiol 1994; 139: 1139-1152. 37. Forastiere F, Agabiti N, Corbo GM et al. Passive smoking as a determinant of bronchial responsiveness in children. Am J Respir Crit Care Med 1994; 149: 365-370. 38. Frischer T, Kuhr J, Meinek R et al. Influence of maternal smoking on variability of peak expiratory flow rate in school children. Chest 1993; 104: 1133-l 137. 39. Rona RJ, Chinn S. Lung function, respiratory illness, and passive smoking in British primary school children. Thorax 1993; 48: 21-25. 40. Schmitzberger R, Rhomberg K, Buchele H et al. Effects of air pollution on the respiratory tract of children. Pediatr Pulmonol 1993; 15: 68-74. 41. Smyth A, O’Hea U, Williams G, Smyth R, Heaf D. Passive smoking and impaired lung function in cystic fibrosis. Arch Dis Child 1994; 71: 353-354. 42. Kitchen WH, Olinsky A, Doyle LW et al. Respiratory health and lung function in g-year old children of very low birthweight: a cohort study. Pediatrics 1992; 89: 1151-1158. 43. Lebowitz MD, Sherrill D, Holberg CJ. Effects of passive smoking on lung growth in children. Pediatr Pulmonol 1992; 12: 3742. 44. Sherrill DL, Martinez FD, Lebowitz MD et al. Longitudinal effects of passive smoking on pulmonary function in New Zealand children. Am Rev Respir Dis 1992; 145: 1136-1141.
DOMESTIC
45. Goren AI, Hellman S. Passivesmoking among school children in Israel. Env Health Perspect 1991; 96: 203-2 11. 46. Guneser S, Atici A, Alparslan N, Cinaz P. Effects of indoor environmental factors on respiratory systemsof children. J Tropical Pediatr 1994; 40: 114-l 17. 47. Willers S, Attewell R, Bensryd I, Schutz A, Skarping G, Vahter M. Exposure to environmental tobacco smoke in the household and urinary cotinine excretion, heavy metals retention, and lung function. Arch Environ Health
1992; 41: 357-363.
48. Hibbert ME, Courlej JM, Landau Li. Changes in lung, airway and chest wall function in boys and girls between 8 and 12 years. J Appl Physiol 1984; 57: 304-308. 49. Vedal S, Schenker MB, Samet JM et al. Risk factors for childhood respiratory disease: Analysis of pulmonary function. Am Rev Respir Dis 1984; 130: 187-192. 50. Becklake MR, Lalloo UG. The ‘healthy smoker’: a phenomenon of health selection? Respiration 1990; 57: 137-144. 51. Weiss ST, Tosteson TD, Segal MR et al. Effects of asthma on pulmonary function in children. A longitudinal population based study. Am Rev Respir Dis 1992; 1456: 58-64. 52. Melia RJW, Florey C du V, Chinn S. Relation between respiratory illness in primary school children and the use of gas for cooking. Results from a National study. Int J Epidemiol 1979; 8: 3333338. 53. Infante-Rivard C. Childhood asthma and indoor environmental risk factors. Am J Epidemiol 1993; 137: 834844.
FACTORSAND
LUNGFUNCTIONINCHILDREN
935
54. Yarnell JWG, St. Leger AS. Housing conditions, respiratory illness and lung function in children in South Wales. Br J Prev Sot Med 1977; 31: 1833188. 55. Willoughby DA. Asthma increase, summer 1994 (letter to the editor). Lancet 1994; 344: 413. 56. Colloff MJ, Ayres J, Carswell F et al. The control of allergens of dust mites and domestic pets: a position paper. Clin Exp Allergy 1992; 22: l-28. 57. Anderson LJ, Parker RA, Strikas RA et al. Day-care center attendance and hospitalization for lower respiratory tract illness. Pediatrics 1988; 82: 300-308. 58. Marbury MC, Maldonado G, Waller L. Lower respiratory illness, recurrent wheezing, and day care attendance. Am J Respir Crit Care Med 1997; 155: 156-161. 59. Becklake MR. Epidemiology of test failure. Br J Znd Med 1990; 47: 73374. 60. Ng’Ang’a LW, Ernst P, Jaakkola MS, Gerardi G, Hanley JH, Becklake MR. Spirometric lung function. Distribution and determinants of test failure in a young adult population. Am Rev Respir Dis 1992; 145: 48-52. 61. Vollmer WM, Johnson LR, McCamant LE, Buist AS. Methodologic issues in the analysis of lung function data. J Chron Dis 1987; 40: 1013-1023. 62. Manga P. Income and access to medical care in Canada. In: Coburn D, D’Arcy C, New P et al., eds. Health and Canadian Society. Ontario: Fitzhenry & Whiteside, Markham, 1981: 325-342. 63. Barker DJP, Godfrey KM, Fall C et al. Relation of birthweight and childhood respiratory infection to adult lung function and death from chronic obstructive airways disease. Br Med J 1992; 303: 671-675.