Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002
To describe trends in the occurrence of the common cold during the first 13 years of life among children who attended different childcare settings early in life.
The Tucson Children's Respiratory Study involves 1246 children enrolled at birth and followed up prospectively since May 1980 through October 1984. Children with data regarding day care use during the first 3 years of life were included in this investigation (n = 991). Parents reported the occurrence of frequent (≥4) colds during the past year by questionnaire when each child was 2, 3, 6, 8, 11, and 13 years of age. Child care at home (no unrelated children), at small day care (1-5 unrelated children), or at large day care (≥6 unrelated children) was reported retrospectively by parental questionnaire when the children were approximately 6 years old.
After adjusting for potential confounding variables, compared with children at home those in large day care had more frequent colds at year 2 (odds ratio [OR], 1.9, 95% confidence interval [CI], 1.0-3.4; P = .04), less frequent colds at years 6 (OR, 0.3, 95% CI, 0.1-0.9; P = .02) through 11 (OR, 0.4, 95% CI, 0.1-1.2; P = .09), and the same odds of frequent colds at year 13 (OR,1.0, 95% CI, 0.3-3.8; P = .95). In addition, compared with children in large day care for 1 year or less those attending large day care for more than 2 years had more frequent colds at year 2 (OR, 1.7, 95% CI, 1.0-3.0; P = .04), less frequent colds at years 6 (OR, 0.5, 95% CI, 0.2-1.1; P = .08), 8 (OR, 0.2, 95% CI, 0.1-1.0; P = .04), and 11 (OR, 0.3, 95% CI, 0.1-1.0; P = .05); and the same odds of frequent colds at year 13 (OR, 0.9, 95% CI, 0.3-2.9; P = .80).
Attendance at large day care was associated with more common colds during the preschool years. However, it was found to protect against the common cold during the early school years, presumably through acquired immunity. This protection waned by 13 years of age.
VIRAL UPPER respiratory tract infections account for approximately 50% of all illnesses and approximately 75% of illnesses in young infants.1,2 Although usually self-limited, the common cold results in approximately 23 million days of school absence, 25 million days of work absence, and 17 million physician visits in the United States annually.3,4 Several studies have shown that children attending day care experience more respiratory illnesses than those cared for at home.5- 9 Given that the use of organized day care and nursery schools has increased, involving 65% of 4-year-old children in 1995, it is important to assess further both the short- and long-term implications of day care use on the health of children.10
Some have speculated that immunity acquired from the increased number of infectious illnesses experienced in day care might protect the child on entrance to primary school.11,12 Indeed, one well-distributed parent information sheet regarding frequent infections in early childhood asks parents to view colds as an educational experience for their child's immune system.13 However, little evidence is available to support this conjecture. In a national survey of children attending day care, those with more previous time in day care did have fewer respiratory tract illnesses than those with less previous time in day care.6 In contrast, 2 studies1,9 found no evidence for a protective effect among children who entered preschool at different ages.
Previous studies might not have detected evidence for protection due to immunity acquired from previous upper respiratory tract infections because they focused on short-term protection during the preschool years. Since more than 100 serotypes of different viruses cause the com mon cold and the serotypes circulating within a community vary from year to year, it is unlikely that acquired immunity gained one year would offer significant protection during the subsequent cold season.14- 16 In addition, antibody-related protection against rhinovirus, the most frequent cause of the common cold, was not evident in children younger than 10 years.16 One retrospective study cited in the medical literature found that 10-year-old children who had attended day care experienced fewer respiratory tract illnesses than those who had not attended day care.17 Therefore, studies involving older children would be helpful in clarifying the relationship between early attendance in day care and the subsequent occurrence of upper respiratory tract infections during childhood.
The Tucson Children's Respiratory Study has prospectively followed up a large cohort of children from birth through 13 years of age. This analysis describes trends in the occurrence of the common cold among these children who attended different child care settings early in life.
A total of 1246 normal newborns were enrolled in the Tucson Children's Respiratory Study between May 1980 and October 1984.18,19 Detailed information on the enrollment process and study design has been published elsewhere.19 Information about the mother's level of education, ethnicity, household pets (dogs or cats), and the number of children younger than 18 years living in the household at the time of the child's birth was obtained from responses on questionnaires administered to parents shortly after their child's birth. Because 95% of the household children younger than 18 years were siblings of the study child, they are referred to here as "siblings." Information about breastfeeding status was obtained from 2 sources: prospectively from data gathered at health supervision visits and retrospectively from responses on a follow-up questionnaire.20,21
The study was approved by the Human Subjects Committee of the University of Arizona, Tucson. Written informed consent was obtained from parents at the time of enrollment and at each in-depth evaluation (which took place at years 6 and 11 of the study).
The parents of the enrolled children completed questionnaires related to their child's respiratory status at years 2, 3, 6, 8, 11, and 13 of the study (mean [SD] age of children, 1.6 [0.8] years, 2.9 [0.9] years, 6.3 [0.9] years, 8.6 [0.7] years, 10.9 [0.6] years, and 13.5 [0.6] years, respectively). At each survey parents were asked: "During the past year, how many head colds (common colds) did this child have?" Available responses at years 2 and 3 were "few" (0-3), "some" (4-5), "frequent" (6-9), or "constant" (>9). At years 6 through 13 available responses were "none, few" (1-3), some (4-5), frequent (6-9), or constant (>9 or continuously). Years 2, 3, 8, and 13 surveys were mailed to respondents. The year 6 and year 11 surveys were completed by parents in conjunction with a more detailed evaluation of respiratory health of the children and their parents. While rates of upper respiratory tract illness have varied considerably among children in different cohorts,22,23 for this study the primary outcome variable, frequent colds, was defined as 4 or more colds during the previous year.
Information about attendance in day care and the number of unrelated children present in the day care setting during the first 3 years of life was obtained for 991 children by a parental questionnaire when the mean (SD) age of the children was 6.7 (1.5) years.8 For this study day care categories were defined as follows: home care (no unrelated children), small day care (1-5 other unrelated children), and large day care (≥6 unrelated children).
Day care "exposure" was considered with reference to both the size of the day care attended and the duration of attendance in a large day care. For the first analysis, which aimed at evaluating the effect of day care size, only the 606 children who were continuously enrolled in the same type of day care from 6 months through 3 years of age were included. This was done to provide analytic clarity and to minimize possible bias due to the migration of children between types of day care due to illness. The second analysis, which aimed at evaluating the effect of duration of attendance in large day care, included all children for whom day care information was obtained.
Data were also obtained at each survey regarding parental smoking habits. A child was considered exposed to household smoking if either parent reported that he or she was currently smoking when the child was 2, 3, 6, or 8 years old. At the time of the year 11 survey, 556 of the children with day care data had a private interview with the study nurse regarding their personal smoking habits. Therefore, at year 11 any child who reported smoking or who had a parent who reported current smoking was considered exposed to household smoking. At year 13 no data were available regarding the child's personal smoking habits and, therefore, their smoking habits at year 11 were used as a proxy measure. Therefore, any child who reported smoking at year 11 or who had a parent who reported current smoking when the child was 13 years old was considered exposed to household smoking at year 13. The odds ratio (OR) of frequent colds at year 13 did not vary when only current parental smoking habits at year 13 (and not child smoking status) were entered in the multivariate model.
Parents were asked if their child had hay fever or any other condition that made his or her nose runny, stuffy, or itchy apart from colds. Children were considered to have allergic rhinitis at years 2 or 3 if parents reported that the child had ever experienced such symptoms. At years 6, 8, 11, and 13 only children reported to have had such symptoms during the previous year were considered as having allergic rhinitis.
Because both duration of attendance in day care and the size of the day care can influence the rates of respiratory tract illness, analyses were completed to best evaluate these 2 factors. Initially, the rates of illness among children attending the 3 sizes of day care (home, small, and large) were compared at each year. To be consistent with previously published studies these groups were limited to children who remained in the same-sized day care from 6 months until 3 years of age.5 The second analysis evaluated the rates of illness among children with varying length of time in a large day care. χ2 Tests were used to compare prevalences of frequent colds between groups and Mantel-Haenszel tests to analyze trends across categories of day care use.
In multivariate analyses the following variables were considered for confounding: sex, maternal education (<16 years vs ≥16 years), number of children in the household younger than 18 years, breastfeeding (<4 months vs ≥4 months), ethnicity (both parents Anglo vs either or neither parent Anglo), presence of pets (dogs or cats) in the household, smoking exposure at home at each survey year or by the child at 11 years of age, and the presence of allergic rhinitis in the child at each survey year. Only those potential confounders associated with the child care size or duration with a P<.10 were entered into the multivariate analyses at each survey year.
To assess for possible participation bias and potential effect modification, the multivariate analyses were initially stratified for each covariate that significantly differed between subjects with and without child care information. Since no effect modification was apparent, the analyses shown herein include the total sample. Analysis was completed with SPSS (SPSS Inc, Chicago, Ill) for UNIX, version 4.0. An α level of .05 (2-tailed) was considered statistically significant, unless noted otherwise.
Data regarding child care during the first 3 years of life were obtained for 991 of 1246 children in the original cohort and, therefore, these children were included in the analysis regarding duration of attendance in large day care. Children excluded from this analysis because of missing data were significantly more likely to be non-Anglo, have a mother with a lower level of education, and have allergic rhinitis at years 6 and 11 than children included in this analysis. In addition, excluded children were significantly less likely to be breastfed for 4 months or longer or to have pets in the household (data not shown). Of these 991 children, 606 were cared for in the same type of child care from 6 months through 3 years of age and were included in the analysis regarding the size of the day care. Children excluded from this phase of the analysis were significantly more likely to be non-Anglo, have mothers with a lower level of education, have no older sibling, and have allergic rhinitis at year 3 than children included in the analysis (data not shown).
Because children included in these analyses differed from those excluded owing to missing data for these characteristics, further analyses were conducted to investigate possible participation bias. Similar trends were seen for each subgroup when these analyses were stratified by each of the factors significantly associated with choice of child care.
The occurrence of frequent colds at each year was compared with the size of the day care among children who were cared for in the same setting for the first 3 years of life (Table 1). The size of the day care was directly proportional to the likelihood of having frequent colds at years 2 (trend χ2, P = .003) and 3 (trend χ2, P = .01). However, children who had attended large day care were less likely than children in either of the smaller day care settings to have frequent colds at years 6 (12% vs 21%, P = .12), 8 (2% vs 17%, P = .005), and 11 (5% vs 17%, P = .02). By year 13 each group of children, regardless of previous attendance in day care, had similar rates of frequent colds. Although children in small day care had more frequent colds at years 2 and 3 than those at home, the 2 groups did not significantly differ in their risk of frequent colds from years 6 through 13.
The relationship observed between frequent colds and child care size was also investigated in multivariate analyses. The ORs of frequent colds for children cared for in large day care were compared with children cared for at home at each year, adjusted for the following potential confounders: ethnicity, maternal educational level, number of older siblings, breastfeeding, and the presence of allergic rhinitis and household smoking at each given year (Table 2). Compared with children at home, those who attended large day care were at greater risk of more frequent colds at year 2 (OR, 1.9; 95% CI, 1.0-3.4; P = .04), less risk of frequent colds at years 6 (OR, 0.3; 95% CI, 0.1-0.9; P = .02) through year 11 (OR, 0.4; 95% CI, 0.1-1.2; P = .09), and at the same risk by year 13 (OR, 1.0; 95% CI, 0.3-3.8; P = .95). The adjusted ORs at each year for frequent colds did not statistically significantly differ between children who attended small day care compared with those who remained at home.
To further explore this relationship between frequent colds and attendance in large day care among all children with day care use data, the duration of time in large day care was compared with the occurrence of frequent colds at each year (Table 3). A trend similar to that seen for day care size was noted, with length of time in large day care being proportional to the risk of frequent colds at years 2 and 3, inversely related from year 6 through year 11, and unrelated at year 13.
The relationship observed between frequent colds and the years spent in large day care was also investigated in multivariate analyses. The ORs of frequent colds according to duration of time spent in large day care were adjusted for ethnicity and the presence of allergic rhinitis (Table 4). Compared with children in large day care for less than 1 year, those who attended large day care for 2 years or longer were at statistically significantly greater risk of frequent colds at year 2 (OR, 1.7; 95% CI, 1.0-3.0; P = .04), significantly less risk of frequent colds at years 8 (OR, 0.2; 95% CI, 0.1-1.0; P = .04) and 11 (OR, 0.3; 95% CI, 0.1-1.0; P = .05), and at similar risk of frequent colds by year 13 (OR, 0.9; 95% CI, 0.3-2.9; P = .80). From years 6 through 11 the risk of frequent colds for children with more than 1 to 2 years of attendance in large day care was intermediate between children with 1 year or less and more than 2 years of attendance, suggesting a dose response by years in large day care. The effect of more than 1 to 2 years of attendance in large day care was significant at year 11 (OR, 0.4; 95% CI, 0.2-1.0; P = .05). The risk of frequent colds at years 2 and 3 is difficult to interpret for children with more than 1 to 2 years in a large day care because, as categorized by duration of attendance, it is unclear actually when the children were in the large day care.
The primary finding of this study was that children who attended large day care during the first 3 years of life had more frequent colds during the preschool years, but less frequent colds during the school years until 13 years of age. In addition, a dose response was noted for years spent in large day care. No such protective effect during the school-aged years was found for attendance in small day care.
Clinicians have often attempted to reassure parents that the frequent upper respiratory tract infections experienced by their child while in day care would strengthen their immune system and thereby offer some protection against infectious diseases after elementary school enrollment. It has been argued that such a protective role of day care, through acquired partial immunity, should be held open as a theoretical possibility until appropriate longitudinal studies were conducted.11
This investigation supports the premise that long-term acquired immunity plays a role protecting against future common colds during childhood. Neutralizing antibodies in serum or nasal secretions have been associated with serotype-specific protection from rhinovirus in adults.24- 26 However, this might not be the case in children as serum antibody–related protection against rhinovirus infection was only evident in family members older than 10 years.16 This could explain why protection owing to attendance in day care was limited to children who attended large day care. Although children in small day care did have more frequent colds during the preschool years, they did not gain any apparent protection during school age in this population. In a young child it is possible that a substantial "booster effect," such as would occur in group day care, might be required to produce adequate and lasting immunity. Alternatively, large day care may increase the number of serotypes to which immunity is acquired by preschoolers.
The results of previous studies regarding the subsequent protection against upper respiratory tract infections from attendance in day care possibly reached different conclusions owing to their inclusion of only very young children and their short-term periods of follow-up, specifically 1 to 2 years.1,5,9 Studies that included children older than 5 years and considered exposure at day care of more than 2 years did find evidence for protection from previous attendance in day care.6,17 Our results confirm these earlier findings obtained from retrospective studies.
The protection acquired from early attendance in day care seems to continue longer than what might have been previously suspected. This is likely explained by the ease of transmission of viruses among preschoolers compared with school-aged children and the seasonal cycling of serotypes within a community. Since toddlers and preschool-aged children are ideal transmitters of infectious agents, they are able to cross-infect each other more efficiently than school-aged children, who are more hygienic and less likely to be ill when infected with rhinovirus.15 In addition, virus serotypes circulating within a community might not reappear for some years. During a 4-year study of respiratory tract illness in Seattle, Wash, Fox et al16 found that of 96 serotypes of rhinovirus, 18 were present in all 4 years, 26 were found in 3 of the years, 24 in 2 of the years, 19 in only 1 of the years, and 9 in none of the years. This is complicated further by the age-specificity of the large number of infectious agents, in addition to rhinovirus, that are responsible for the common cold syndrome. Therefore, if a child were to acquire immunity against a given serotype it might be some years before he or she would have the good fortune to demonstrate adequate protection during a subsequent encounter with the same virus serotype.
There are 2 potential weaknesses of this study. Parental reporting of illnesses might have been inaccurate. Unfortunately, few objective assessments are available for measuring the common cold in children. Previously published reports concerning childhood illness use parents' perceived deviation from normal as their standard.5,6 This study used the same measure previously used in the medical literature9 and benefited from prospective reporting by the same parents for the duration of the study. Colds occurring among attendees of day care might have been more memorable, especially if they led to exclusion from day care or interruption of a parent's work schedule. However, while the type of day care might have influenced parental recall during the preschool years, it is unlikely to have had any influence on recall from years 6 through 13. In addition, the findings for years 2 and 3 are consistent with previous studies investigating the relationship between attendance in day care and upper respiratory tract infections.5 Although loss of follow up was small in this study compared with other longitudinal studies of this duration, it may introduce a potential source of participation bias. However, no evidence for such a bias was found; the observed trends persisted when examined within specific subgroups, making it unlikely that missing data affected these results.
The results of this study are consistent with the hypothesis that long-term acquired immunity obtained in day care is important for the subsequent protection of children against the common cold. Regardless of whether children acquire that immunity in preschool or primary school, they seem to have similar levels of protection by 13 years of age. Clinicians may continue to reassure parents of children in day care that their child's plight with minor respiratory tract illnesses is not in vain.
Accepted for publication September 7, 2001.
This work was supported by grants HL 14136 and HL 56177 from the National Heart, Lung, and Blood Institute Specialized Center of Research, Rockville, Md.
Presented in part at the 38th Annual Meeting of the Ambulatory Pediatric Association, New Orleans, La, May 3, 1998.
We would like to acknowledge the participation of the study families and the work of the study nurses, Marilyn A. Smith, RN, and Lydia L. De La Ossa, RN. Our appreciation to Bruce W. Saul, MS, for his assistance with data analysis and Kathryn M. Ball for her review of the manuscript.
Children in day care experience more infectious illnesses than children cared for at home. The results of studies investigating whether these infections protect attendees of day care from future illnesses were conflicting. This study examined trends in upper respiratory tract illnesses among a cohort of children followed up prospectively from birth through 13 years of age. Children in large day care were found to have more frequent colds during the preschool years but less frequent colds during the early school years. This apparent protection waned by 13 years of age.
Corresponding author: Thomas M. Ball, MD, MPH, Department of Pediatrics, University of Arizona College of Medicine, 1501 N Campbell Ave, PO Box 245073, Tucson, AZ 85724-5073 (e-mail: email@example.com).
Ball TM, Holberg CJ, Aldous MB, Martinez FD, Wright AL. Influence of Attendance at Day Care on the Common Cold From Birth Through 13 Years of Age. Arch Pediatr Adolesc Med. 2002;156(2):121-126. doi:10.1001/archpedi.156.2.121