Objective. To examine the relationship between day care attendance and illnesses of the upper and lower respiratory tract in the first year of life.
Study Design. Prospective birth cohort study.
Methods. Children (N = 498) who had at least 1 parent with a history of allergy or asthma were enrolled at birth and followed prospectively for the first year of life. A home visit at 2 to 3 months of age and bimonthly telephone questionnaires were used to obtain information on day care arrangements, home characteristics, respiratory symptoms, and physician-diagnosed illnesses of the upper and lower respiratory tract.
Results. Day care attendance in the first year of life was associated with two or more doctor-diagnosed ear infections (OR: 2.4; 95% CI: 1.7–3.6), three or more parental reports of runny or stuffed nose (OR: 3.2; 95% CI: 1.9–5.5), a doctor's diagnosis of sinusitis (OR: 2.2; 95% CI: 1.1–4.2), and doctor-diagnosed lower respiratory illnesses (croup, bronchitis, bronchiolitis, and pneumonia; OR: 1.6; 95% CI: 1.0–2.4). For children attending day care, exposure to pets in day care, the presence of a rug or carpet in the area where the child slept in day care, and a nonresidential setting for day care all were independent predictors of two or more doctor-diagnosed ear infections.
Conclusions. The results suggest that day care increases the risk of illnesses of the upper and lower respiratory tract in the first year of life for children with a familial history of atopy. Specific environmental exposures within day care, such as the presence of pets or having a rug or carpet in the area where children sleep, may increase the risk of recurrent ear infections in the first year of life among children with familial history of atopy who attend day care.
- LRI =
- lower respiratory illness •
- OR =
- odds ratio •
- URI =
- upper respiratory illness •
- CI =
- 95% confidence interval
Day care attendance by children, particularly those <2 years of age, is associated with increased prevalence and severity of certain upper and lower respiratory tract infections.1–7Of the children in the United States, ∼60% attend some form of day care.8
Children with parental history of atopy are at higher risk for developing diseases of the upper and lower respiratory tract than are children born to nonatopic parents.9 ,10 Little is known, however, about the effects of day care in young children at high risk for atopy. If day care attendance in early life was associated with an increased risk for respiratory illnesses in this population, then identifying modifiable risk factors within day care would be important in terms of both public health and clinical practice.
The Home Allergens and Asthma Study is a prospective birth cohort study of children with a parental history of asthma/allergies in the Boston metropolitan area. The primary purpose of the study is to assess the relationship between exposure to indoor allergens in early childhood and the subsequent development of bronchial asthma and atopy. As part of this study, information has been collected on day care to test the hypothesis that sources of allergens and respiratory irritants within day care contribute to the risk of respiratory infections, wheeze, or allergic symptoms in childhood. In this report, we describe the association of day care attendance with symptoms or illnesses of the upper and lower respiratory tract in the first year of life, and we examine whether specific exposures in day care can explain this association.
The 505 infants (including six sets of twins) and their 499 families with a history of allergy or asthma in at least 1 parent were recruited between September 1994 and August 1996. The screening and recruitment of families have been described elsewhere.10In brief, eligibility criteria included residence inside route 128 (a highway encircling the Boston metropolitan area); maternal age ≥18 years; history of hay fever, bronchial asthma, or allergies in either parent; and maternal ability to speak English or Spanish. Families were not screened if the newborn was hospitalized in the intensive care unit, if his or her gestational age was <36 weeks, or if he or she had a congenital anomaly. A screening telephone interview was conducted when the child was 1 month of age, before his or her mother returned to work. After obtaining written informed consent, a home visit was made when the child was 2 to 3 months of age, and a questionnaire regarding home characteristics, environmental exposures, smoking, and demographics was administered by trained research assistants. Every 2 months, beginning when the child was 2 months of age, a follow-up telephone questionnaire was administered to the child's primary caretaker. Information on respiratory symptoms and illnesses experienced by the index child since the previous telephone interview, day care arrangements, infant feeding, and selected home characteristics was recorded. The relationship between the day care environment and respiratory symptoms and illnesses was assessed using information from the home visit and the bimonthly telephone questionnaires. Of the children, 7 were excluded from analysis because they were followed for ≤4 months during their first year of life.
Definition of Predictor Variables in Day Care
Day care was treated as a categorical variable (yes, if the children attended any type of day care in the first year of life; and no, otherwise). The type of day care was classified as home (at the day care provider's residence), nonresidential (not provided at someone else's home), and mixed. The number of hours per day, the number of hours per week, and the number of months per year that the index child attended day care were treated as both continuous and categorical variables.
Potential sources of allergens or respiratory irritants in day care were considered. The index child was categorized as exposed to pets in day care if there were dogs, cats, hamsters or gerbils, pet mice or rats, other furry pets, or birds. The presence of a rug or wall-to-wall carpeting in the area where the index child slept in day care was coded as a categorical variable. Finally, exposure to smoking in day care was treated as both a categorical and a continuous variable.
Definition of Other Predictor Variables
Potential confounders of the relationship between day care and respiratory illnesses were considered. Sociodemographic variables included the child's race or ethnicity (according to parental report), as previously described,10 and annual household income, classified as low (<$30 000), middle ($30 000–$49 999), or high (≥$50 000). Perinatal and familial factors included in utero exposure to smoking (yes or no); postnatal exposure to maternal smoking (average number of cigarettes per day smoked by the mother inside the household); number of months of breastfeeding (0–2, 3–8, or 9–12); bottle-feeding in a bed or crib before a nap or sleep time; parental history of asthma (mother, father, or both), and parental history of hay fever (mother, father, or both). Variables related to the home environment included number of older siblings (≤14 years of age) in the household; total cigarette smoking (average number of cigarettes per day smoked by all adults in the household); presence of a rug or carpet in the area where the child slept; and presence of any pets.
Definition of Outcome Variables
Every 2 months the primary caretaker was asked, “Since we last spoke with you on [date given], has your child had an ear infection diagnosed by a doctor?” Other bimonthly questions included whether any of the following had occurred since the previous questionnaire was administered: a runny or stuffed nose; doctor-diagnosed sinus trouble; doctor-diagnosed lower respiratory illness (LRI; croup, bronchitis, bronchiolitis, or pneumonia); and wheezing or whistling in the chest. All the outcome variables were categorized a priori according to previously published studies11 and potential clinical relevance, as follows: two or more versus fewer than two doctor-diagnosed ear infections; three or more versus fewer than three reports of runny or stuffed nose (nasal catarrh); any versus no doctor-diagnosed sinusitis; any versus no LRIs; any versus no reports of wheeze; and two or more versus fewer than two reports of wheeze.
SAS statistical software (SAS Institute, Cary, NC) was used to evaluate univariate and multivariate associations between the predictor variables and the outcomes of interest. Univariate associations between categoric predictor variables and outcomes were examined with 2 × 2 contingency tables. We used logistic regression to study the relationship between day care attendance and outcomes while controlling for potential confounders and examining interactions. Stepwise logistic regression was used to develop the multivariate models. In the final models, we included those variables that satisfied a change-in-estimate criterion (≥10% in the odds ratio [OR] estimate) or that were significant at the P < .05 level.
Characteristics of Cohort and Day Care Experience
Table 1 summarizes the primary characteristics of the 498 subjects. Of the 238 children (47.8%) who attended day care in their first year of life, 161 (67.7%) went to a home setting, 52 (21.8%) to a nonresidential setting, and 25 (10.5%) to both (mixed day care). Of the 50 children (21.0%) who attended day care with at least 10 other children, 38 (76%) attended nonresidential day care exclusively in their first year of life. All the 238 children who attended day care in their first year of life did so for ≥5 hours per week and for ≥1 month. Of the 238 children, 109 (45.8%) attended day care for ≥6 months, ≥3 days per week, and ≥4 hours per day.
Association of Day Care with Respiratory Illnesses
Table 2 shows the results of the analysis of the association between day care attendance in the first year of life and respiratory illnesses.
Upper Respiratory Illnesses (URI)
After adjustment for breastfeeding and number of older siblings in the household, children who attended day care in the first year of life had 2.4 times higher odds of two or more doctor-diagnosed ear infections than did children who did not attend day care (Table 2). In a multivariate analysis, living with at least 1 older sibling in the household was an independent predictor of two or more doctor-diagnosed ear infections in the first year of life (OR: 2.8; 95% confidence interval [CI]: 1.8–4.1), as was breastfeeding (OR for 9–12 months of breastfeeding: 0.6; 95% CI: 0.4–0.96; P < .05).
After adjustment for the number of older siblings in the household, children attending day care in the first year of life had 3.2 times higher odds of three or more parentally reported episodes of nasal catarrh than did children not attending day care (Table 2). In addition, the presence of at least 2 older siblings in the household was associated with increased odds of this outcome (OR: 2.5; 95% CI: 1.1–5.5).
The results of the adjusted analysis of the relationship between day care and doctor-diagnosed sinusitis were not significantly different from those of the univariate analysis; children who attended day care in the first year of life had 2.2 times higher odds of doctor-diagnosed sinusitis than did children who did not attend day care (Table 2).
Household income, child's race and gender, in utero smoke exposure, postnatal exposure to maternal smoking, total cigarette smoking in the household, and parental history of asthma or hay fever were not associated independently with any URI in a multivariate analysis. Furthermore, we found no significant modification of the effect of day care on URI by any of these variables.
LRI and Wheezing
After adjustment for gender and number of older siblings in the household, children attending day care in the first year of life had 1.6 times higher odds of doctor-diagnosed LRIs (croup, bronchitis, bronchiolitis, and pneumonia) than did those not attending day care (P < .05; Table 2). In a multivariate analysis, the number of older siblings in the household was an independent predictor of doctor-diagnosed LRIs (OR for living with 1 sibling: 2.0; 95% CI: 1.2–3.1; OR for living with 2 or more siblings: 2.6, 95% CI: 1.5–4.6), as was male gender (OR for boys versus girls: 2.0; 95% CI: 1.6–2.5). Exposure to cigarette smoking (prenatal or postnatal), parental history of asthma or hay fever, child's ethnicity, and household income were not predictors of doctor-diagnosed LRIs in a multivariate analysis. Furthermore, there was no modification of the effect of day care on doctor-diagnosed LRIs by any of these variables.
After adjustment for LRIs, bottle-feeding in the bed or crib before a nap or sleep, cigarette smoking by all adults in the household, and maternal history of asthma, children attending day care in the first year of life had a statistically nonsignificant trend toward higher odds of having any wheezing (P = .11) and two or more reports of wheezing (P = .24) than did children who did not go to day care (Table 2). In a multivariate analysis, bottle-feeding the child in the bed or crib before a nap or sleep was an independent predictor of higher odds of any wheeze report (OR: 1.8; 95% CI: 1.2–2.8) and of two or more reports of wheezing (OR: 3.2; 95% CI: 1.8–5.4). In addition, total cigarette smoking by adults in the household was associated with higher odds of recurrent wheezing (OR for two or more reports of wheezing for every 10 additional cigarettes: 3.2; 95% CI: 1.3–7.5).
Multivariate Analysis Restricted to Children in Day Care
To explore the relationship between potential risk factors within day care and respiratory illnesses, we conducted a multivariate analysis restricted to the 238 children attending day care in their first year of life (Table 3).
After adjustment for number of siblings in the household, number of children in day care, presence of pets in day care, presence of a rug or carpet in day care, and breastfeeding, children attending nonresidential day care in their first year of life had 2.9 times higher odds of two or more doctor-diagnosed ear infections than did children attending day care at someone else's home or mixed day care.
In a multivariate analysis, children attending day care in settings with at least 10 other children had 2.2 times higher odds of two or more doctor-diagnosed ear infections in the first year of life (P = .10) than did children attending day care in settings with fewer than 10 children. Furthermore, the presence of any pet in day care was associated with 1.9 times higher odds and the presence of a rug or carpet in the area where the child slept with 3.2 times higher odds of this outcome.
The frequency with which children attended day care was not associated independently with recurrent ear infections in a multivariate analysis. Although we did not find an increased risk for two or more doctor-diagnosed ear infections associated with passive smoking in day care, the fact that only 29 (12.2%) of the 238 children reportedly were exposed to smoking by day care providers may have reduced our power to detect a difference between the groups.
The only factor within day care that was associated significantly with three or more episodes of nasal catarrh was attending day care with at least 3 other children (OR: 3.9; 95% CI: 1.6–10.0). For the 238 children attending day care, neither exposures within day care nor frequency of day care attendance was associated significantly with doctor-diagnosed sinusitis, LRIs, any wheeze report, or two or more reports of wheezing.
We demonstrated an association between day care attendance in the first year of life and recurrent ear infections, recurrent nasal catarrh, doctor-diagnosed sinusitis, and doctor-diagnosed LRIs among children with parental history of atopy. Furthermore, we explored the relationship between exposures within day care and respiratory diseases in the first year of life controlling for exposures in the child's household and other confounders. Nonresidential day care, exposure to pets in day care, and the presence of a rug or carpet in the area where the child slept in day care all were associated with higher odds of recurrent ear infections in the first year of life. In addition, attending day care with at least 3 other children was associated with increased odds of recurrent nasal catarrh in the first year of life.
Previous studies of community-dwelling children that have been conducted in different countries, have involved children of different socioeconomic status, and that have used different methodologies and outcome definitions have consistently found risks of at least twofold for otitis media among young children attending day care.3 ,6 ,11 In a recent prospective cohort study, 2253 children living in urban, suburban, and small town or rural settings in the Pittsburgh, PA, area were followed monthly, and a diagnosis of acute otitis media was made by pneumatic otoscopy and tympanometry.12 Subjects living in urban areas were more likely to be black children on Medicaid, whereas children in suburban and small town or rural areas were more likely to be white and to have private health insurance. Regardless of the study site, subjects attending day care with other children had a significantly greater mean cumulative percentage of days with middle ear effusion in their first year of life than did subjects who were cared for in the absence of other children. Furthermore, the investigators, Paradise and associates, described a strong association between otitis media in the first year of life and the degree of exposure to other children, whether at home or in day care. They postulated that exposure to a large number of children was more important than was the setting in which the exposure took place. However, we found that the type of day care (nonresidential vs home/mixed) was an independent predictor of recurrent ear infections. Therefore, in children with a parental history of allergy/asthma, nonresidential day care may be an important risk factor for recurrent ear infections in early childhood.
In our study, the presence of any pet or of a rug or carpet in areas where children slept in day care was predictive of recurrent ear infections. Because children in our cohort had a parental history of atopy/asthma, they may have been at unusually high risk for upper airway inflammation on exposure to allergens present in animals or carpets in day care. Alternatively, fomites (rugs and pets) may contribute to the spread of viral infections among children attending day care. Inflammation of the nasal mucosa attributable to allergy or infection has been associated with eustachian tube dysfunction leading to persistent middle ear effusions13 and is probably associated with an elevated risk of otitis media as well.
The observed association between day care and recurrent nasal catarrh may reflect an increased risk not only of infection but also of allergy. In a longitudinal study, rhinitis in the first year of life was predictive of allergic rhinitis and asthma at age 6; this result suggested that rhinitis in early life is a manifestation of an atopic predisposition.9 Other authors also have suggested that nasopharyngeal viral infections or edema of the nasal mucosa attributable to allergy may predispose children to sinusitis.14
In a cohort study in which day care exposure was assessed retrospectively, Holberg et al15 demonstrated higher odds of LRIs in the first year of life among infants exposed to at least 3 unrelated children attending day care (OR for children 6 to 12 months: 1.7; 95% CI: 1.1–2.5).
In a recent analysis of data from the Indoor Air and Children's Health Study, Marbury et al4 demonstrated that day care attendance was associated with a threefold increase in the odds of having three or more episodes of wheezing-associated LRIs before 2 years of age. In contrast, we found only a modest increase in the odds of having two or more reports of wheezing among children attending day care in their first year of life. We found no significant associations between sources of allergens in day care (rug or carpet or pets) and recurrent wheezing. However, we followed children only for their first year of life, when most episodes of wheezing reflect inflammation of the small airways attributable to LRIs.16–18 In addition, reporting of allergen sources may be a weak surrogate for true allergen exposure.19
Bottle-feeding a child before a nap or sleep was associated with an almost threefold increase in the odds of ≥2 reports of wheezing. Bottle-feeding has been associated with a higher risk of recurrent wheezing after bronchiolitis attributable to parainfluenza virus.20
We found no association between bottle-feeding before a nap or sleep and LRIs. Rather than increasing the risk of infection, bottle-feeding the child in the supine position may cause inadvertent aspiration leading to bronchoconstriction and subsequent wheezing.
Like other scholars,4 ,12 ,15 we found no significant association between frequency of day care attendance in the first year of life and illnesses of the upper or lower respiratory tract. Thus, among children with familial history of atopy, exposure to day care in the first year of life for ≥5 hours per week during ≥1 month is associated with an increased risk of URI and LRI.
We recognize several limitations to our findings. First, our study was not designed to be a random sample of the population in the greater Boston area, because we selected a stable population with parental history of allergy or asthma. It should be pointed out, however, that the predilection of study participants for home-based day care reflects the general experience in the greater Boston area in which ∼76% of licensed day care is provided in a home setting (Office of Child Care Services for Massachusetts, personal communication, 1999).
Second, if children were required to be examined by a physician to return to day care after any illness, there could be ascertainment bias for physician- diagnosed illnesses. To the best of our knowledge, Massachusetts does not have a policy requiring that parents bring a physician's note after illnesses such as an ear infection or an LRI (Day Care Licensing Regulations and Policies, Office of Child Care Services for Massachusetts); random calls to 10 home-based day care centers and 10 nonresidential day care centers in our study confirmed that this was the case. More important, studies conducted in different countries and using different methods for outcome ascertainment have found ORs similar to ours for the effect of day care on recurrent ear infections and LRIs. Thus, we do not believe that ascertainment bias is a likely explanation for our findings. Third, there was probably some misclassification of both day care and disease status. However, misclassification of exposure and outcome were likely independent of each other (nondifferential), the result being an underestimation of the true ORs. Finally, we assessed the day care environment by parental report, because visiting all the centers would have been prohibitively expensive. Nevertheless, our study is unique in that we collected extensive and frequent information on the day care and home environments.
In summary, we found an association between day care attendance in the first year of life and illnesses of the upper and lower respiratory tract in a population of children at high risk for atopy. In addition, we identified modifiable factors within day care that are associated with an increased risk for recurrent ear infections among children with parental history of atopy. Longitudinal follow-up of our birth cohort should provide significant insight as to whether day care attendance in early life has a protective or harmful effect on the development of asthma and allergic rhinitis in childhood.
This study was supported by Grant AIEHS35786 from the National Institutes of Health. Dr Celedon is supported by National Research Service Award Grant HL07427.
- Received September 16, 1998.
- Accepted April 1, 1999.
Reprint requests to (J.C.C.) Channing Laboratory, Brigham and Women's Hospital, 181 Longwood Ave, Boston, MA 02115. E-mail:
- Schwartz B,
- Giebink S,
- Henderson FW,
- Reichler MR,
- Jereb J,
- Collet JP
- Anderson LJ,
- Parker RA,
- Strikas RA,
- et al.
- Collet JP,
- Burtin P,
- Kramer MS,
- Floret D,
- Bossard N,
- Ducruet T
- Wald ER,
- Guerra N,
- Byers C
- ↵Dawson DA, Cain VS. Child care arrangements: health of our nation's children, United States, 1988. Adv Data Vital Health Stat. October 1, 1990;187. DHHS Publication PHS 90-1250
- Wright AL,
- Holberg CJ,
- Martinez FD,
- Halonen M,
- Morgan W,
- Taussig LM
- Paradise JL,
- Rockette HE,
- Colborn K,
- et al.
- Holberg CJ,
- Wright AL,
- Martinez FD,
- Morgan WJ,
- Taussig LM,
- Group Health Medical Associates
- Copyright © 1999 American Academy of Pediatrics