PEDIATRICS Vol. 118 No. 5 November 2006, pp. e1369-e1380 (doi:10.1542/peds.2005-2345)
ARTICLE |
Compliance With American Academy of Pediatrics and American Public Health Association Illness Exclusion Guidelines for Child Care Centers in Maryland: Who Follows Them and When?
a Division of General and Community Pediatrics Research, Cincinnati Children's Hospital Medical Center, and Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio
b Housestaff Department, Children's Hospital of Pittsburgh, Pittsburgh, Pennsylvania
c Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
d Division of General Pediatrics and Adolescent Medicine, Johns Hopkins University, Baltimore, Maryland
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
BACKGROUND. In 1992, the American Academy of Pediatrics and the American Public Health Association jointly published guidelines for temporary exclusion of sick children from child care. However, little is known about key stakeholders' compliance with these guidelines.
OBJECTIVES. The purpose of this work was to compare pediatricians', parents', and child care providers' compliance with American Academy of Pediatrics guidelines and determine predictors for higher rates of compliance.
METHODS. We conducted a cross-sectional survey of 215 randomly selected Maryland pediatricians, 223 parents, and 192 child care providers from 22 Baltimore, Maryland, child care centers from January to July 2004. Questionnaires contained the following 6 case vignettes depicting common child care illnesses: upper respiratory infection, conjunctivitis, gastroenteritis, mild febrile illness, tinea capitis, and atopic dermatitis. The instrument measured the correctness of exclusion and inclusion decisions (using American Academy of Pediatrics/American Public Health Association guidelines as gold standard) according to varying levels of fever, disease severity (eg, clear versus yellow eye discharge), familiarity with the child, and parent work schedule flexibility.
RESULTS. Response rates were 71% for pediatricians, 56% for parents, and 85% for child care providers. Guideline compliance was higher for pediatricians (74%) than for child care providers (60%) and parents (61%). Only 23% of pediatricians and parents and 29% of child care providers reported familiarity with American Academy of Pediatrics/American Public Health Association guidelines by name. In general, child care providers and parents had lower false-negative rates (allowed fewer children to attend who met criteria for exclusion) than pediatricians, suggesting that pediatricians may underexclude. Child care providers and parents correctly excluded in 65%–98% of cases requiring exclusion, whereas pediatricians correctly excluded 31%–86% of cases requiring exclusion, depending on the vignette. Yet pediatricians were much more specific about which children met criteria (pediatricians correctly included 61%–93% of cases requiring inclusion versus child care providers and parents who correctly included 20%–75% of such cases), suggesting that child care providers and parents may overexclude. Compliance rates varied significantly by stakeholder, vignette (disease), level of fever, and disease severity but did not vary with the stakeholder's familiarity with the child or the flexibility of the parent's work schedule.
CONCLUSIONS. Pediatricians, parents, and child care providers were unfamiliar with American Academy of Pediatrics/American Public Health Association illness exclusion guidelines by name but moderately compliant with them. When noncompliant, child care providers and parents generally overexcluded, and pediatricians underexcluded. Stakeholder- and disease-specific predictors for noncompliance gleaned from this study suggest how educational interventions aiming to increase guideline compliance could be individually tailored to child care providers, parents, and pediatricians.
Key Words: exclusion • infection control • child care • health policy
Abbreviations: AAP—American Academy of Pediatrics • APHA—American Public Health Association • CCP—child care provider • URI—upper respiratory infection
In 2001, 61% of US children aged 0 to 5 years were in some form of nonparental care. More than 50% of 3- to 6-year-olds and just under 20% of 0- to 2-year-olds attended centers, including preschools.1 Children in center care are at increased risk for respiratory and gastrointestinal illnesses.2–7 In the interest of the health and safety of children who become ill at the center, as well as the other exposed healthy children, individual centers construct exclusion policies that dictate which children must be sent home or temporarily "excluded" from child care. These exclusion policies must comply with state licensing requirements. Many states' policies (including Maryland's8) have remained imprecise over the last 20 years about which symptoms mandate exclusion.9
With little direction from the state, it is not surprising that previous studies10–14 of child care exclusion policies found widely varying policies among individual centers. Mindful of these reports, experts in early education, public health, and infectious disease from the American Academy of Pediatrics (AAP), the American Public Health Association (APHA), and the Maternal and Child Health Bureau convened in 199215 and again in 200216 to develop consensus guidelines for health and safety in child care. Among these guidelines were recommendations for the temporary exclusion of ill children from child care settings, which were based on the diverse experts' assessment of children's needs, caregiver abilities, public health risks, and the best available evidence for infection control.17 The guidelines use language understandable to laypersons18 to define 3 overall conditions for exclusion: (1) inability of the child to participate in program activities, (2) the caregiver's inability to provide care for the child without compromise of care for the other children in the group, and (3) a list of 28 specific symptoms and diseases that warrant temporary exclusion from child care, and 7 symptoms and diseases that do not warrant exclusion. These guidelines are widely available in books,16,19 booklets,20,21 and on the Internet.16
Since their release, few studies have examined to what extent child care providers (CCPs) are aware of these guidelines and/or use them in their daily exclusion decisions.14,22–24 None of these studies used standardized vignettes to compare practices. Moreover, although parents and pediatricians make important contributions to the exclusion decision process10,25 (parents may retain their ill child at home before they can be excluded,22 and pediatricians may be asked to write readmission notes to child care), we are aware of no published studies regarding to what extent parents' and pediatricians' practices or recommendations comply with AAP/APHA guidelines.
Only 2 studies11,25 simultaneously compared all 3 groups of stakeholders. Landis et al11 published in 1988 that CCPs excluded more often than pediatricians for the same symptoms, yet the study preceded release of the guidelines by several years. A more recent study25 found CCPs, parents, and pediatricians' equally knowledgeable about general (but not necessarily AAP/APHA) exclusion guidelines. However, this study did not examine compliance or which factors might be associated with better guideline compliance. Both of these studies used symptom lists to test respondents, thus ignoring the potentially important role of social context26 in medical and layperson exclusion decision-making.
This study had 2 aims: (1) to compare CCP, parent, and pediatrician compliance with AAP/APHA guidelines using case vignettes; and (2) to determine demographic and situational factors (including awareness and familiarity with published guidelines by title) associated with better guideline compliance. We studied familiarity with the published guidelines, because lack of awareness and familiarity have been identified as barriers to physician guideline adherence.27 However, we recognized that exclusion decisions may align with the guidelines even if respondents were unaware of the guidelines by name. Using standardized clinical vignettes in which 2 clinical and 2 social situational factors (hereafter referred to collectively as the 4 "scenario variables") varied randomly, we aimed to test the sensitivity and specificity of the stakeholders to the symptom thresholds for exclusion per the guidelines, as well as to 2 social contextual factors: child familiarity and parent work schedule flexibility.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study Design and Population
In this cross-sectional survey, we surveyed the following 3 stakeholder groups: parents, caregivers in Baltimore child care centers, and Maryland pediatricians.
Child Care Centers
The Maryland Child Care Administration provided a comprehensive list of all of the licensed child care centers in Baltimore County and Baltimore City as of July 2003 (N = 627). Centers were randomly selected (using a random-number generator) from 4 strata based on location (city versus county) and size (
42 vs >42 child capacity) and contacted by telephone to verify eligibility and begin recruitment. Facilities located in private residences (n = 3), providing only partial day care (n = 38), serving exclusively children >5 years (n = 7), listing no information about capacity (n = 4), unreachable by telephone (n = 5), or that closed during the course of the study (n = 2) were excluded. Recruited directors estimated by telephone the number of caregivers (including themselves) involved in daily exclusion decisions in their center. Each center was allotted a number of parent surveys such that the number of collected parent surveys would roughly equal the number of collected staff surveys per center, assuming 50% and 85% response rates for parents and CCPs, respectively.
All of the CCPs, directors, and administrative staff identified by the 22 participating center directors as participating in exclusion decisions were eligible and selected to participate, regardless of their possible absence on the day of survey distribution. All of the parents with 
1 enrolled child under the age of 5 years were eligible; only 1 survey response per household was permitted.
Pediatricians
All of the pediatricians who had completed residency training, were AAP chapter members, and who identified themselves as regularly seeing and making decisions about Maryland children excluded from child care (per instructions on first page of the questionnaire) were eligible to participate. Five hundred pediatricians were randomly selected from a list provided by the Maryland chapter of the AAP (N = 1121). The sampling frame for pediatricians was extended from Baltimore to Maryland to obtain adequate sample size (n = 200 for each stakeholder group) to detect with 80% power a 15% difference in compliance rates between groups (
= .05).
Survey Instrument
Each questionnaire contained 6 case vignettes representing common child care-associated illnesses (Table 1). The vignettes were generated using a standardized clinical description of the condition and random permutations of 4 scenario variables that were hypothesized as affecting respondent's exclusion answers: (1) 2 levels of respondent familiarity with child (familiar vs unfamiliar or own child vs a classmate for parents); (2) 3 levels of fever (normal, low-grade, and below the 100°F AAP/APHA threshold for "fever,"16 or high fever >101.5°F); (3) 2 levels of parental work schedule flexibility (flexible vs inflexible); and (4) 2 levels of disease severity (except tinea capitis; see Table 1). (High fevers were not assigned to the tinea capitis or atopic dermatitis case vignettes.) The measures of disease severity were chosen because they were explicitly listed in the AAP/APHA guidelines. Regardless the level of each of the 4 scenario variables that was randomly assigned to each of the 6 vignettes, every survey had 6 correct answers (either to exclude or not exclude) based on the guidelines corresponding to the 6 vignettes (see Table 1). Alignment with AAP/APHA guidelines16 of the respondent's 6 answers to: "would you recommend this child be sent home now?" (options: "yes" or "no") comprised our main outcomes of interest (compliance). Because the assignments are random, this study design can be used to investigate the independent effect of potentially sensitive scenario variables28,29 on the outcome of exclusion compliance, holding other variables constant.
|
We assessed respondents' familiarity with local center sick policies, state licensing regulations, and AAP/APHA guidelines (identified by both title and contributing organizations) with a 5-point Likert scale. To examine possible social desirability bias in self-reported familiarity, a phony set of guidelines entitled "American Coalition of States' Exclusion Guidelines for Child Care Centers" was listed among the set of existent guidelines.
The questionnaire also collected respondents' demographic data including experience with child care. To enhance content validity, the questionnaire was pretested for clarity and term-appropriateness among 20 CCPs, parents, and pediatricians.
Survey Administration
Questionnaires were collected between January and July 2004. The self-administered, anonymous questionnaire took, on average, between 15 minutes for pediatricians and 20 minutes for CCPs and parents to complete. An accompanying cover letter described the study's general intent, risks and benefits, and right of participant refusal. This study was approved by the Johns Hopkins Medicine Institutional Review Board.
Questionnaires were distributed to all of the staff participating in exclusion decisions and returned to the investigators on a subsequent visit or by mail. Most parents (n = 355; 18 centers) were randomly selected for participation (usually by selecting every jth name on a class list). At 2 small centers (n = 17), all of the parents were selected; a small number of parents (n = 33; 2 centers) were approached for participation consecutively at child pick-up time.
Analysis
An overall compliance rate was calculated as the percentage of the 6 vignettes in which self-reported decisions were congruent with AAP/APHA guidelines (possible range: 0%–100%; higher scores indicating better compliance). Mean compliance rates for centers were calculated separately for center parents and CCPs. As a covariate hypothesized as relating to compliance, respondent's guideline familiarity was dichotomized into those indicating they were "very familiar" or "familiar" with guidelines versus all other responses. Respondents indicating that they were "very familiar" with the phony set of guidelines and highly familiar with all of the other guidelines listed (n = 34) were removed from the familiarity analysis to partially control for presumed social desirability bias. Compliance rates and guideline familiarity among groups were compared using analysis of variance and 
2 test, where appropriate.
For each vignette, we calculated the sensitivity and specificity of each stakeholder group's exclusion decisions compared with AAP/APHA guidelines as the gold standard using the following formulas:
Sensitivity [TP/(TP+FN)] = (number of correct exclusions by vignette-group)/(total number of exclusions recommended by AAP/APHA guidelines)
Specificity [TN/(FP+TN)] = (number of correct inclusions by vignette-group)/(total number of inclusions recommended by AAP/APHA guidelines)
To determine predictors for better compliance, we conducted bivariable and multivariable linear regressions of overall compliance rate stratified by stakeholder group on the following demographic variables, all selected a priori: self-reported familiarity with local, state, and AAP/APHA guidelines; educational attainment; race/ethnicity; having children (yes/no); and previous personal experience with child care. Robust SEs for CCP and parent compliance rates were generated using Stata's "cluster" command to adjust for clustering by center.30 All of the statistically significant bivariable predictors (as well as education regardless of its statistical significance if race/ethnicity was a significant bivariable predictor) were added to multivariable linear regression models.
The remaining analyses were stratified by vignette and stakeholder group, with SEs adjusted for clustering by center when appropriate. We compared within a stakeholder group the effect of altering each of our scenario variables (level of disease severity, fever, familiarity with child, and parental work schedule flexibility) on the outcome of compliance. The test of statistical significance was based on multivariable logistic regression; compliance (yes/no) was the dependent variable, and the scenario variables were the independent variables. For clinical scenario variables, we calculated adjusted probabilities of compliance from our stratified multivariable logistic regression results, setting all of other scenario variables equal to their mean. Statistical analysis was performed using Stata 8.0 (Stata Corp, College Station, TX). For all of the analyses, P < .05 was considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sample Characteristics and Response Rate
Of the 500 pediatricians selected to participate, 353 (71%) responded. Of these, 138 (39%) stated that they were ineligible because they were not actively practicing general pediatrics in the state of Maryland, leaving a final sample size of 215 pediatricians. Of the 226 eligible CCPs, 192 (85%) responded; 223 (56%) of the 401 eligible parents that we surveyed responded.
Demographic data regarding our sample is presented in Table 2. Parents' surveys were most often completed by the mother (89%). Eighty percent of parents reported working full-time, 13% part-time, and 6% reported not working. Among the CCPs, 21% reporting working at the center <1 year and 45% <3 years.
|
Compliance
The majority of CCPs, parents, and pediatricians indicated exclusion decisions consistent with AAP/APHA guidelines, except for tinea capitis (Table 3). Pediatricians were more likely than parents and CCPs to comply with AAP/APHA guidelines for upper respiratory infection (URI), conjunctivitis, and tinea capitis. However, the 3 groups had indistinguishable compliance rates for gastroenteritis, mild febrile illness, and atopic dermatitis. Mean center compliance rates ranged widely across and within centers (Table 3).
|
Sensitivity and Specificity of Exclusion Decisions
Because respondents' noncompliant exclusion decisions could be inappropriate exclusions or inclusions, we calculated the sensitivity and specificity of exclusion decisions by group and disease type (vignette; Table 4). Overall, CCPs and parents had higher sensitivities (lower false-negative rates) to AAP/APHA thresholds for exclusion than pediatricians, suggesting that pediatricians may at times underexclude. Conversely, pediatricians' recommendations were much more specific than parents' and CCPs' (lower false-positive rates), indicating that parents and CCPs may at times overexclude. Pediatricians had high false-negative rates (sensitivity = 31%) for children who were miserable from their atopic dermatitis and required exclusion. Parents' and CCPs' relatively low specificities for tinea capitis (37% and 21%, respectively) and gastroenteritis (41% and 31%, respectively) indicated overexclusion for children who did not yet meet criteria. This imbalance between sensitivities and specificities indicates that CCPs, parent, and pediatricians all could not determine the exact threshold for exclusion versus inclusion among varying presentations of 6 common child care-associated illnesses.
|
Familiarity
We found that only about one fifth of all 3 of the groups indicated familiarity with AAP/APHA exclusion guidelines by name (Table 5). On the other hand, CCPs and parents indicated more familiarity with local (75% and 68%, respectively) and state (54% and 24%, respectively) exclusion guidelines.
|
Demographic Predictors of Compliance
Pediatricians who were familiar with a local child care center policy, had used a child care center for their own children, or were relatively junior, had slightly higher (2%–6%) compliance rates; however, these associations were no longer statistically significant after adjusting for all of the hypothesized covariates (see "Methods" section). After adjusting for educational attainment, income, clustering of center, and other statistically significant bivariable predictors, parents and CCPs who reported their race as black had slightly lower (7%) compliance rates. CCPs with more years of experience had slightly higher compliance rates (test for trend: P < .01) after adjusting for other factors.
Scenario Predictors of Compliance
Using stratified logistic regression, we examined within a stakeholder group the effect of altering each of our scenario variables (familiarity with child, parental work schedule flexibility, level of disease severity, and fever) on the outcome of compliance, holding other scenario variables constant. Neither familiarity with the child nor parental work schedule flexibility significantly altered compliance rates for any of the stakeholder groups in any of the diseases. Therefore, the remainder of the analysis will focus on level of fever and disease severity.
Figures 1 through 3 illustrate that compliance was significantly affected by which temperature was presented to the respondents in most cases. For instance in Fig 1, among pediatricians, higher temperatures were inversely associated with compliance, as pediatricians presented with a child with URI symptoms and no fever had a 93% adjusted probability of compliance with guidelines, whereas a child with URI symptoms and a high fever garnered a 68% adjusted probability of compliance among pediatricians. Conversely, parents' (Fig 2) and CCPs' (Fig 3) compliance rates demonstrated U- and J-shaped patterns, with the "low-grade fever" (temperature = 99.5°) associated with lowest compliance rates (47% and 41%, respectively) in a child with URI. The way in which temperature affected compliance varied by stakeholder and disease. If a scenario variable did not significantly alter compliance within a vignette and stakeholder group (P > .05 in Figs 1–6), it suggests that the stakeholder is attuned to the guidelines' recommended exclusion threshold for that variable.
|
, afebrile;
, low-grade temp <100°; 
, high fever >101.5°. a Adjusted probabilities were calculated from stratified multivariable logistic regression results by setting all other scenario variables to their mean value.
|
|
|
Compliance depended on each symptom tested among CCPs (Fig 6), on 2 of the 5 symptoms tested among pediatricians (Fig 4), and on 3 of the 5 symptoms tested among parents (Fig 5), suggesting limited awareness of the guidelines' symptom thresholds for exclusion. Specifically, the child with URI and green (vs clear) nasal discharge and the child with "gastroenteritis" who had only vomited once (vs twice) had lower compliance rates for parents and CCPs. The child with (allergic) conjunctivitis and clear eye discharge had lower compliance rates among all 3 of the groups (Figs 4–6). In a child with atopic dermatitis, the activity level altered compliance rates for pediatricians (Fig 4) and CCPs (Fig 6), albeit in opposite directions. Pediatricians were more likely to fail to exclude a child miserable from her pruritus, and CCPs were more likely to fail to include a child with a benign rash.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Most CCPs, parents, and pediatricians were unfamiliar with AAP/APHA guidelines by name. Compliance rates, however, indicated moderate adherence to these guidelines in all 3 of the groups, despite a lack of familiarity. On the whole, pediatricians better complied with guidelines (74% of vignettes) than parents (61%) or CCPs (60%). Compliance rates for parents and CCPs were similar to knowledge levels of national guidelines reported by a separate study of a similar population.25 Pediatricians scored slightly better in this study, which used vignettes, compared with the previous study,25 which used symptom lists. As would be expected from other studies,10–14,31,32 compliance rates by center varied considerably.
When examined collectively, none of the stakeholders were highly sensitive and specific to the AAP/APHA symptom thresholds for exclusion. CCPs and parents missed fewer appropriate opportunities for exclusion (lower false-negative rates) for mildly ill children than pediatricians, but when paired with their high false-positive rates (specificities ranged 20%–75% for parents and 21%–63% for CCPs), this high-sensitivity/low-specificity pattern suggested that parents and CCPs overexclude.
We found several significant clinical predictors of poorer compliance. When compared with the extremes, the low-grade fever proved to be the most difficult for parents and CCPs in the setting of URI, conjunctivitis, and gastroenteritis. For pediatricians, lower compliance was evident in cases of URI with high fever, conjunctivitis with clear discharge, high fever in an otherwise asymptomatic child, and a child with miserably pruritic atopic dermatitis.
Reported familiarity with national guidelines and familiarity with the child did not predict compliance with national guidelines after adjusting for other factors. Contrary to our hypothesis, we did not find stakeholders "bending the rules" for children whose parents' had inflexible work schedules.
There was a persistent association of reported black race with compliance, even after adjusting for education and income. This association may reflect unmeasured confounding perhaps related to shared health beliefs of a particular ethnic and cultural group.33
Methodologic Considerations
There are several potential limitations to this study. First, we measured self-reported exclusion decisions; we cannot know how decision-makers would respond in real situations. However, vignettes have been reported to be equivalent to standardized patients and less expensive and superior to chart review in measuring physician behavior.34,35 Second, we only allowed our respondents 2 options, to exclude or include, because we were interested in measuring their personal inclinations before any other stakeholders began to weigh in. In reality, the exclusion outcome is more likely to be a result of a negotiation between CCP and parent or parent and physician. Third, our reported guideline familiarity may underrepresent true levels of familiarity, because respondents may have been familiar with guidelines' content but not name, and our fictitious guidelines may have been mistaken for existent guidelines. This may especially be true for parents and CCPs with little administrative experience, none of whom would be expected to be familiar with guidelines' titles. With respect to predictors of compliance, because overall compliance rates were moderately high, we were not powered to examine predictors of overexclusion versus overinclusion, although it is very possible that they may differ. Fourth, we had to adopt strict interpretations of the AAP/APHA guidelines to determine a correct answer for each of the 6 vignettes. Although these guidelines are more clear and specific than many state guidelines, there is still room for the decision-maker's subjective judgment, as in 3.065b, which states, "any illness that results in a greater need for care than the child care staff can provide without compromising the health and safety of the other children as determined by the child care provider" requires exclusion. Therefore, our measurements of compliance were interpretations of the criteria. The guidelines state that, "children who cannot use a toilet for all bowel movements while attending the facility and who develop diarrhea should be removed from the facility" (6.023a).
Diarrhea is not defined for non–toilet-trained children, but for staff it is defined by "3 or more episodes of diarrhea in the previous 24 hours or blood in the stools" (3.069). For toilet-trained children, excludable diarrhea is defined as "more watery stools, decreased form of stool that is not associated with changes in diet, and increased frequency of passing stool, that is not contained by the child's ability to use the toilet" (3.065c3). Some respondents may have considered our "2 loose stools contained by the diaper" as excludable by these criteria. Considering this, we performed a sensitivity analysis such that the correct answer to the gastroenteritis vignette was always to exclude. Although this changed the compliance rates for the gastroenteritis vignette (pediatricians' compliance dropped to 64%, and parents' and CCPs' compliance improved to 75% and 83% respectively), the overall compliance rates (73%, 62%, and 62%, respectively) did not change appreciably, nor did the demographic predictors for overall compliance.
Finally, our findings may not be generalizable to other populations, especially those with different state guidelines. Our sample included a large portion of black CCPs and parents, yet studies have shown that this group uses formalized child care more often than Hispanic or white race/ethnicities.36 Although our response rates were similar or higher than other studies of similar populations,11,23,37,38 it is possible that those 44% of parents and 15% of CCPs who chose not to participate may have systematically differed in their exclusion decisions than our respondents. This study was neither designed nor powered to examine the effect of center variables (eg, for-profit versus nonprofit and rural versus urban location) or caregiver variables (eg, age group cared for or teacher versus administrator) on exclusion decisions. We surveyed only AAP members, assuming that they would be more familiar with AAP/APHA guidelines given the AAP's involvement in their writing. Non-AAP members may or may not be more familiar and compliant with AAP/APHA guidelines.
Nevertheless, we believe that our findings have several important potential implications. The variability in compliance with exclusion guidelines by disease, stakeholder, and disease severity suggests opportunities for tailored educational interventions. Recent studies estimate that respiratory, gastrointestinal and febrile illnesses and rashes account for >75% of illnesses and absent days for children in child care39,40 and that children experience the associated symptoms as often as 23% of the time.41 Good epidemiological prevalence estimates are lacking regarding how often these 6 conditions occur and meet exclusion criteria. Assuming an average absence of 1.25 days per excludable illness episode39 and an incidence of 6.6 excludable illness episodes per child per year,40 our measured CCP overall specificity of 43% (a weighted average based on relative vignette prevalence39,40) suggests that for every child appropriately excluded, 6 children are inappropriately excluded. Improving CCPs' specificity to pediatrician levels (75%) would reduce the number of inappropriate exclusions to 2.6 children per 1 child appropriately excluded. On the other hand, underexclusions (high false-negative rates or low sensitivities) among CCPs and parents could delay necessary medical evaluation and treatment and reduce the quality of child care for other children due to the special demands of an ill child. Of greater concern among pediatricians, low sensitivities to public health risks, such as infectious diarrhea among diapered children in a group setting, could foster premature readmissions to child care, increased disease spread, and outbreaks of potentially serious pathogens, such as Escherichia coli 0157:H7.
Given the prevalence of child care-related illness and associated symptoms, educating all of the stakeholders about the 3 conditions for exclusion, namely: (1) child's ability to participate, (2) caregiver's ability to care for the child without compromise of care for other children in the group, and (3) fever and symptom thresholds for exclusion related to the risk of disease transmission within the group, could lead to more appropriate exclusions and less conflict among the views of parents, CCPs, and pediatricians. Educating all 3 of the groups about symptom thresholds, what temperature level constitutes an excludable fever, and that fevers must be accompanied by an associated symptom to warrant exclusion could potentially reduce many unnecessary exclusions, as well as inappropriate and possibly harmful inclusions/readmissions to the child care setting.
As with any diagnostic test, the decision to exclude a child involves a delicate tension between the decision-makers' sensitivity for appropriately excluding potentially ominous and excludable symptoms and their specificity for appropriately including benign symptoms. Children deemed eligible for exclusion by the initial decision-makers (CCPs and parents) with relatively high sensitivities and brought to a second tier of decision-makers (pediatricians or child care directors) with high specificities resembles a 2-stepped screening test for rare but potentially serious diseases, such as HIV.42 Yet, our findings suggest that pediatricians' (the second level of testing) specificities are far from optimal (only half exceeded 70%). Also, as with false-positive screening results, unnecessary exclusions that result from overly sensitive and underspecific first-level decision-makers (CCPs) can cause undue stress for parents in terms of work loss. Moreover, many parents may not have access or may not choose to consult their pediatrician for a child unnecessarily excluded, thus amplifying the effect of the initial CCP's high false-positive rate. Our findings highlight the many potential gains of improving each stakeholder's specificity with respect to AAP/APHA exclusion recommendations.
Because >50% of CCPs in our study were familiar with state guidelines, one possible way to improve familiarity with the content of AAP/APHA guidelines would be to adopt them into individual state licensing criteria. Many states require that licensed centers have exclusion policies but offer no specific suggestions for symptom criteria.9 Alternatively, because >70% of parents and CCPs indicated familiarity with center guidelines, adopting them at the center level might greatly increase familiarity and compliance within that center. AAP/APHA guidelines could also be adopted by child care accrediting agencies, such as the National Association for the Education of Young Children. A final method to improve compliance is wider use of trained child care health consultants,43 who can foster a 2-way learning dialogue between caregivers in the community and medically trained health professionals. Still, because these guidelines have not been universally adopted, and because their adoption in selected centers has not been studied, it remains to be seen whether adoption of AAP/APHA guidelines can successfully decrease the spread of child care-associated illness or reduce unnecessary child care and work absences.
|
ACKNOWLEDGMENTS |
|---|
This study was funded by the Robert Wood Johnson Clinical Scholars Program.
We thank Neil Powe, MD, and Marie Diener-West, PhD, for their assistance in this project's design and implementation and Nancy Kovacs for her expertise in early education. We are indebted to Greg Flatt, MS, and Steve Johns, MS, for their assistance in producing the questionnaires. We are grateful for Tarik Najeddine and Leslie McHugh and their tireless help with data entry. Finally we thank all of the CCPs, parents, and pediatricians who participated in this study.
|
FOOTNOTES |
|---|
Accepted Jun 5, 2006.
Address correspondence to Kristen A. Copeland, MD, General and Community Pediatrics Division, Cincinnati Children’s Hospital Medical Center, MLC 7035, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail: kristen.copeland{at}cchmc.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- US Federal Interagency Forum on Child and Family Statistics. America's children in brief 2004. Available at: www.childstats.gov/ac2004. Accessed July 18, 2005
- Churchill RB, Pickering LK. Infection control challenges in child-care centers. Infect Dis Clin North Am. 1997;11 :347 –365[CrossRef][Web of Science][Medline]
- Wald ER, Guerra N, Byers C. Frequency and severity of infections in day care: three-year follow-up. J Pediatr. 1991;118 :509 –514.[CrossRef][Web of Science][Medline]
- National Institute of Child Health and Human Development Early Child Care Research Network. Child care and common communicable illnesses: Results from the National Institute of Child Health and Human Development Study of Early Child Care.
Arch Pediatr Adolesc Med. 2001;155
:481
–488
[Abstract/Free Full Text] - National Institute of Child Health and Human Development Early Child Care Research Network. Child care and common communicable illnesses in children aged 37 to 54 months.
Arch Pediatr Adolesc Med. 2003;157
:196
–200
[Abstract/Free Full Text] - Churchill RB, Pickering LK. Infections in child care centers. Curr Opin Infect Dis. 1996;9 :176 –180[CrossRef][Web of Science]
- Slack-Smith LM, Read AW, Stanley FJ. A prospective study of absence for illness and injury in childcare children. Child Care Health Dev. 2002;28 :487 –94[CrossRef][Web of Science][Medline]
- Electronic Source: Maryland Department of Human Resources. The Child Care Administration, Office of Licensing. Available at: www.dhr.state.md.us/cca/license/index.htm. Accessed January 31, 2003
- National Resource Center for Health and Safety in Child Care. Available at: http://nrc.uchsc.edu/states.html. Accessed January 28, 2006
- McConnochie KM, Wood NE, Kitzman HJ, Herendeen NE, Roy J, Roghmann KJ. Telemedicine reduces absence resulting from illness in urban child care: Evaluation of an innovation.
Pediatrics. 2005;115
:1273
–1282
[Abstract/Free Full Text] - Landis SE, Earp JA, Sharp M. Day-care center exclusion of sick children: comparison of opinions of day-care staff, working mothers, and pediatricians.
Pediatrics. 1988;81
:662
–667
[Abstract/Free Full Text] - Shapiro ED. Exclusion of ill children from day-care centers. Policy and practice in New Haven, Connecticut.
Clin Pediatr (Phila). 1984;23
:689
–691
[Abstract/Free Full Text] - Chouillet A, Maguire H, Kurtz Z. Policies for control of communicable disease in day care centres.
Arch Dis Child. 1992;67
:1103
–1106
[Abstract/Free Full Text] - Addiss DG, Sacks JJ, Kresnow MJ, O'Neil J, Ryan GW. The compliance of licensed US child care centers with national health and safety performance standards.
Am J Public Health. 1994;84
:1161
–1164
[Abstract/Free Full Text] - Aronson SS. The science behind the American Public Health Association/American Academy of Pediatrics National Health and Safety Guidelines for Child-Care Programs.
Pediatrics. 1994;94
:1101
–1104.
[Abstract/Free Full Text] - American Academy of Pediatrics, American Public Health Association, National Resource Center for Health and Safety in Child Care, ed. Caring for Our Children: National Health and Safety Performance Standards: Guidelines for Out-of-Home Child Care Programs. 2nd ed. U.S. Department of Health and Human Services Health Resources and Services Administration Maternal and Child Health Bureau; 2002. Available at: http://nrc.uchsc.edu/CFOC/index.html. Accessed January 28, 2006
- Richardson M, Elliman D, Maguire H, Simpson J, Nicoll A. Evidence base of incubation periods, periods of infectiousness and exclusion policies for the control of communicable diseases in schools and preschools. Pediatr Infect Dis J. 2001;20 :380 –391[CrossRef][Web of Science][Medline]
- Chang A. American Public Health Association/American Academy of Pediatrics National Health and Safety Guidelines for Child-Care Programs: An overview.
Pediatrics. 1994;94
:1107
–1108.
[Abstract/Free Full Text] - American Academy of Pediatrics. Red Book: Report of the Committee on Infectious Diseases; 2003. 26th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2003
- American Academy of Pediatrics. Stepping Stones for Using Caring for Our Children. 2nd ed. Elk Grove Village, IL: American Academy of Pediatrics; 2003
- Aronson SS, Shope, TR. Managing Infectious Diseases in Child Care and Schools: A Quick Reference Guide. Elk Grove Village, IL: American Academy of Pediatrics; 2005
- Friedman JF, Lee GM, Kleinman KP, Finkelstein JA. Child care center policies and practices for management of ill children. Ambul Pediatr. 2004;4 :455 –460[CrossRef][Web of Science][Medline]
- Pappas DE, Schwartz RH, Sheridan MJ, Hayden GF. Medical exclusion of sick children from child care centers: A plea for reconciliation. South Med J. 2000;93 :575 –578[Web of Science][Medline]
- Skull SA, Ford-Jones EL, Kulin NA, Einarson TR, Wang EE. Child care center staff contribute to physician visits and pressure for antibiotic prescription.
Arch Pediatr Adolesc Med. 2000;154
:180
–183
[Abstract/Free Full Text] - Copeland KA, Duggan AK, Shope TR. Knowledge and beliefs about guidelines for exclusion of ill children from child care. Ambul Pediatr. 2005;5 :365 –371[CrossRef][Web of Science][Medline]
- Clark JA, Potter DA, McKinlay JB. Bringing social structure back into clinical decision making. Soc Sci Med. 1991;32 :853 –866[CrossRef][Web of Science][Medline]
- Cabana MD, Rand CS, Powe NR, et al. Why don't physicians follow clinical practice guidelines? A framework for improvement.
JAMA. 1999;282
:1458
–1465
[Abstract/Free Full Text] - Kinchen KS, Cooper LA, Wang NY, Levine D, Powe NR. The impact of international medical graduate status on primary care physicians' choice of specialist. Med Care. 2004;42 :747 –755[CrossRef][Web of Science][Medline]
- Thamer M, Hwang W, Fink NE, et al. U.S. nephrologists' attitudes towards renal transplantation: results from a national survey. Transplantation. 2001;71 :281 –288[CrossRef][Web of Science][Medline]
- Stata Corp. Stata Reference Manual: Stata Release 8.0 [computer program]. College Station, TX: Stata Press; 2003
- Shapiro ED, Kuritsky J, Potter J. Policies for the exclusion of ill children from group day care: An unresolved dilemma. Rev Infect Dis. 1986;8 :622 –625[Web of Science][Medline]
- Hurwitz ES, Gunn WJ, Pinsky PF, Schonberger LB. Risk of respiratory illness associated with day-care attendance: A nationwide study.
Pediatrics. 1991;87
:62
–69
[Abstract/Free Full Text] - Helman C. Culture, Health, and Illness. 4th ed. Oxford, United Kingdom and Boston, MA: Butterworth-Heinemann; 2000
- Dresselhaus TR, Peabody JW, Luck J, Bertenthal D. An evaluation of vignettes for predicting variation in the quality of preventive care. J Gen Intern Med. 2004;19 :1013 –1018[Web of Science][Medline]
- Peabody JW, Luck J, Glassman P, Dresselhaus TR, Lee M. Comparison of vignettes, standardized patients, and chart abstraction: A prospective validation study of 3 methods for measuring quality.
JAMA. 2000;283
:1715
–1722
[Abstract/Free Full Text] - US Census Bureau. Who's minding the kids? Child care arrangements: Spring 1999. Available at: www.census.gov/population/www/socdemo/child/ppl-168.html. Accessed December 9, 2004
- Polyzoi E, Babb J. The challenge of caring for mildly ill children: a Canadian national childcare study. Early Child Res Q. 2004;19 :431 –448[Web of Science]
- Friedman JF, Lee GM, Kleinman KP, Finkelstein JA. Acute care and antibiotic seeking for upper respiratory tract infections for children in day care: Parental knowledge and day care center policies.
Arch Pediatr Adolesc Med. 2003;157
:369
–374
[Abstract/Free Full Text] - Cordell RL, MacDonald JK, Solomon SL, Jackson LA, Boase J. Illnesses and absence due to illness among children attending child care facilities in Seattle-King County, Washington.
Pediatrics. 1997;100
:850
–855
[Abstract/Free Full Text] - Cordell RL, Waterman SH, Chang A, Saruwatari M, Brown M, Solomon SL. Provider-reported illness and absence due to illness among children attending child-care homes and centers in San Diego, Calif.
Arch Pediatr Adolesc Med. 1999;153
:275
–280
[Abstract/Free Full Text] - Carabin H, Gyorkos TW, Soto JC, Penrod J, Joseph L, Collet JP. Estimation of direct and indirect costs because of common infections in toddlers attending day care centers.
Pediatrics. 1999;103
:556
–564
[Abstract/Free Full Text] - Centers for Disease Control and Prevention. Revised guidelines for HIV counseling, testing, and referral. MMWR Recomm Rep. 2001;50 :1 –57; quiz CE1–19a1–CE6–19a1[Medline]
- Evers DB. The pediatric nurse's role as health consultant to a child care center. Pediatr Nurs. 2002;28 :231 –235[Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
