 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Department of Pediatrics, National Jewish Medical and Research Center, Denver, Colorado
Department of Psychiatry, University of Colorado Health Sciences Center, Denver, Colorado
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Method. Infants 9 to 24 months old, considered at risk for developing asthma on the basis of having had 3 or more health care contacts with documented wheezing, received comprehensive evaluations as part of an environmental intervention study. Baseline evaluations with the infants, their families, and their home environments focused on biological, environmental, and psychosocial factors that would potentially increase asthma risk for the children. At study entry, prior morbidity attributable to wheezing illness was assessed with caregiver reports of symptom frequency and severity and medication use, caregiver quality of life, and medical record documentation of hospitalizations and emergency department (ED) visits.
Results. Forty-six percent of the infants had 1 or more hospitalizations and 59% had 2 or more ED visits since birth for wheezing illness. Foreign-born Hispanic families had significantly more ED visits for their children’s wheezing illness than US-born Hispanic families, whites, or blacks, although they used fewer controller medications and they reported less illness severity. Multivariate analyses showed 3 biological factors, respiratory syncytial virus, elevated child IgE, and cockroach allergen in the home, were independently associated with hospitalizations within this sample. Similar analyses showed that ED visits were not associated with biological variables, but rather with caregivers with single parent status and smokers. Caregiver reports of wheezing illness severity were correlated with ED visits, but not with hospitalizations. Severity ratings were higher for children of mothers with asthma and for those whose caregivers had higher anxiety and stress. The only correlate of caregiver ratings of poor quality of life was high caregiver anxiety.
Conclusions. Ethnic and immigrant status was an important factor in morbidity attributable to infant wheezing illness. In addition to respiratory infection, both allergic processes and social variables were associated with morbidity as measured by health care utilization. Caregiver reports of illness severity were significantly correlated with psychosocial factors.
Key Words: infant wheezing illness • morbidity patterns • psychosocial factors • environmental allergens • allergy in infants • quality of life • low-income families
Abbreviations: ED, emergency department • CAPS, Childhood Asthma Prevention Study • CQoL, Caregiver Quality of Life • RSV, respiratory syncytial virus • SD, standard deviation • IgE, immunoglobulin E
This study examines the correlates of morbidity attributable to wheezing illness in a multi-ethnic sample of low-income infants younger than age 2 years. Both prevalence and morbidity of asthma and wheezing illness are greater among minority children of lower socioeconomic status.1–3 Recent investigations have provided important information regarding wheezing and asthma among low-income school-aged children.4 However, much less is known about morbidity attributable to wheezing illness or asthma among infants and toddlers, despite epidemiologic evidence suggesting that much of the increase in asthma morbidity among low-income, urban children has occurred in this age group.1,2
Biological factors known to be associated with wheezing in infants, including prematurity, perinatal smoke exposure, low birth weight, and young maternal age2,5–9 are often more prevalent among low-income families. Acute wheezing among infants younger than age 2 has been related primarily to viruses, especially respiratory syncytial virus (RSV), and to passive smoke exposure, measured by salivary cotinine, whereas acute wheezing in children older than age 2 was associated with evidence of allergy.10 However, some studies have reported that early exposure to allergens was associated with infant wheeze.11 The role of allergen exposure for infant wheeze requires further investigation. Wheezing symptoms in the first years of life may signify the initial manifestations of childhood asthma,12,13 although more than half of infants who wheeze before 3 years of age do not go on to develop childhood asthma.13,14 However, the probability of developing persistent asthma goes up with the development of atopy, a history of parental asthma, and increasing frequency of early wheeze.15
Psychosocial factors have also been associated with increased prevalence of wheezing illness and related morbidity of wheezing illness. Poverty, single parent status, and low maternal education have been associated with more wheezing illness among children younger than 5 years of age.2 Maternal mental health problems and lack of social support have been associated with increased emergency department (ED) visits and hospitalizations for school-aged children with asthma.16–18 Information is needed regarding the manner in which psychosocial variables are associated with greater illness severity and health care utilization for infants with repeated wheezing episodes.
The Childhood Asthma Prevention Study (CAPS) provided an opportunity to examine morbidity attributable to wheezing illness among infants and toddlers. For CAPS, infants were recruited on the basis of having had 3 or more health care contacts during which wheezing was documented. The goal of this randomly controlled study was to determine the efficacy of a home-based environmental intervention. At CAPS study entry, extensive evaluations with the infants, their families, and their home environments were conducted to determine additional biological, environmental, or psychosocial factors that would potentially increase asthma risk for these wheezing children. Analyses of the risk factors showed high variability within the sample, with different patterns of risk factors being significantly associated with ethnic status and country of origin.19 The foreign-born Hispanic group was most strikingly different, showing significantly lower infant urinary cotinine levels and fewer reports of maternal asthma than any other group.
An additional goal of the CAPS baseline assessment was to quantify baseline illness severity; therefore, extensive information regarding morbidity attributable to wheezing illness was also obtained. Measuring morbidity attributable to wheezing illness raises particular difficulties for infant and toddler-aged children. The most commonly used indexes of morbidity for older children and adults are hospitalizations and ED visits. Utilization of hospitals and EDs is not only costly, but is a clear indication of uncontrolled illness. Parent reports of symptom frequency and severity are reasonable morbidity measures, but are subject to influence associated with caregivers’ cognitive and emotional dispositions.20 Caregivers’ reports of quality of life are used to measure the impact of illness on families,21 but for older children the relationship between illness events and quality of life has been overshadowed by caregiver anxiety.22
Because of the difficulties inherent in measuring wheezing illness severity among children, and especially among infants and toddlers, all of these measures were used in CAPS. Information was obtained regarding caregiver reports of symptom frequency and severity and of associated caregiver quality of life (CQoL). In addition, complete medical records of health care contacts were obtained and coded regarding hospitalizations and ED visits since the children’s births, providing health care utilization data independent of parent report.
Our objectives in this report are: 1) to describe morbidity attributable to wheezing illness in this low-income group of children using health care utilization and caregiver report measures; 2) to examine infant biological characteristics, environmental exposures, and caregiver psychosocial characteristics in relation to health care utilization and caregiver report morbidity variables; and 3) to examine relationships among the several measures of morbidity with the goal of understanding what each contributes to an overall view of morbidity attributable to wheezing illness in very young children.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Families classified themselves as white (n = 38), black (n = 40), Hispanic (n = 84), Native American (n = 4), or Asian (n = 1). Hispanic families were classified as US-born or foreign-born. Among US-born Hispanic families, 65% spoke English and 35% were bilingual; whereas among foreign-born Hispanic families, 62% spoke only Spanish and 38% were bilingual. Caregivers were primarily infants’ biological mothers (91%), but 7 caregivers (4%) were grandmothers, 4 (2%) were biological fathers, and 5 (3%) had other relationships with the children (eg, foster mothers). Demographic and medical background variables were obtained through standardized interview (birth order, prenatal smoke exposure, maternal education, marital status, maternal age, maternal asthma, asthma or allergy among biological parents or siblings, infant eczema) or medical record review (birth weight, RSV presumed or confirmed).
Biological Assessments
Total serum IgE levels were determined for 144 infants by fluoroenzyme immunoassay. Children were classified as having high IgE according to previous reports for infants who subsequently developed asthma (9–12 months: 3.5 IU/mL; 12–18 months: 5 IU/mL; 18–24 months: 7.5 IU/mL).23,24
Allergen sensitization was assessed for 161 infants by allergy skin-prick testing to common indoor inhalant allergens and food allergens by Quintest (Bayer, Spokane, WA). Aeroallergen extracts used for skin testing were dust mite, cat, dog, cockroach, and mouse; food allergen extracts were milk, egg, and soy; the positive control was histamine, and the negative control was glycerol in saline (Bayer, Elkert, IN). A skin test was regarded as positive if the wheal reaction was 
2 mm in diameter larger than the negative control.25
Environmental Assessments
To measure environmental tobacco smoke exposure, urine samples were obtained for 168 infants and subjected to radioimmunoassay analysis of cotinine, a specific metabolite of nicotine (American Health Foundation Labs, Valhalla, NY). Levels are reported as nanograms of cotinine per gram of creatinine.26
Baseline house dust samples were collected by a research assistant who vacuumed the family living area, the bedroom carpet beside and underneath the bed, the upper surface of the bed, and kitchen floor areas adjoining the cupboard. The combined dust was sieved (>200 mg per household) and was sent for analysis to the Johns Hopkins University DACI Reference Laboratory for determination of allergens, including cockroach (Bla g 1), dust mite (Der f 1 and Der p 1), cat dander (Fel d 1), and dog dander (Can f 1).27–29
Psychosocial Assessments
The Rand Mental Health Battery30 assesses mental health using a list of 38 items that were read to the caregiver during the interview. The measure includes 5 subscales: anxiety, depression, positive affect, emotion, and stability, and a summary mean score. Internal consistency reliability has been shown to be high (
= 0.96).31 The scale has been used effectively with minority populations, and a Spanish version was available. Anxiety scale scores range from 0 to 5; mean scores are reported here.
Caregivers were administered scales (verbal and nonverbal) from the Woodcock-Johnson Tests of Cognitive Ability32 or the Woodcock-Munoz Battery 33: Picture Vocabulary (comprehension and knowledge) and Analysis-Synthesis (fluid reasoning). Recent norms are available for both the English and Spanish versions. Standard scores for the 2 scales were correlated 0.53 (P < .0001) in this sample, and scores were averaged for a mean cognitive score.
A series of 21 questions regarding possible stressful life events in the past year was verbally administered to each caregiver.34 The list included events such as moves, deaths, job losses, court appearances, separations or divorces, or decrements in financial status. These were summed to get a life event score ranging from 0 to 21.
Morbidity: Health Care Utilization
Utilization data, including number of hospitalizations for wheezing illness, number of ED visits, and number of steroid bursts, were obtained from the medical records. Medical records were obtained for every medical contact for each child, including birth records. A standard system developed at the General Clinical Research Center-Psychosocial Assessment Core Laboratory at National Jewish Medical and Research Center35 was used to code the medical encounters. Inter-rater reliability, assessed by a second coder for 25% of the records, ranged from 0.88 to 0.97 (0.62–0.99 95% confidence interval), indicating very good inter-rater reliability.
Morbidity: Caregiver Reports
Caregivers were asked to rate their infants’ symptom frequency for the past 6 months using a 5-point scale: 1) daily, 2) weekly, 3) monthly, 4) less than monthly, or 5) never. The symptoms were 1) wheezing, coughing, or tightness in the chest; 2) nights awakened by breathing problems; 3) slowed or stopped play attributable to breathing problems; 4) breathing problems on awakening. They were asked to respond yes/no to 5) has child had breathing problems so severe that he/she couldn’t vocalize. Scores reported here reflect a mean across the 5 items. This scale was adapted from a parent report asthma severity scale36 developed for school-aged children. Scale reliability was good (Cronbach’s = 0.89) and validity for school-aged children was supported by significant correlations with pulmonary lung function, school days missed, resource utilization, and medication use.36
Caregivers responded to the Pediatric Asthma CQoL scale,21 which consists of 13 items regarding impact of the child’s asthma on the caregiving parent. The CQoL assesses 2 dimensions, caregivers’ emotional response and restriction on family activities. Reproducibility has been found to be good (intraclass correlation = 0.84). For this study, "asthma" items were reworded "when your child experienced cough, wheeze, or breathlessness" or "breathing problems," so that the scale was applicable for infants with wheezing illness.
Data Analyses
Univariate data are presented as means (standard deviation [SD]) where variables were normally distributed and percent of whole for categorical data. Laboratory data with skewed distributions (IgE, urinary cotinine, and cat and cockroach allergen levels) were log transformed for analysis but back transformed for presentation as geometric means (95% confidence interval). For analysis by ethnic groups, normally distributed data were compared using analysis of variance (analysis of variance) procedure. Posthoc analyses were performed using Fischer Least Significant Difference test when the overall analysis of variance was significant, allowing us to find the source of the group difference. Health care utilization data were analyzed in 2 different ways: 1) The data are presented categorically where hospitalizations were cut at 1 or more visits and ED visits at 2 or more visits. The cut of 2 or more ED visits was chosen because only approximately a quarter of the sample had no ED visits, whereas approximately half of the sample had 2 or more. These categories are presented as percent of whole and compared using 
2 analysis. 2) Because of variable time between subjects’ birth and study entry, a ratio variable, "number of events since birth/age in months at study entry," was constructed to adjust for age at study entry. These age-adjusted utilization variables have skewed distributions; therefore, Wilcoxon nonparametric procedure was performed to test group differences.
Spearman correlations were computed to test the relationships between demographic and risk variables with the age-adjusted utilization variables. Spearman partial correlations were then performed to adjust for the confounding effects that the ethnic groups have on the relationships between the risk and the utilization variables. To investigate the simultaneous effects that the univariately significant demographic and risk factors have on health care utilization, multiple logistic regression was performed. In the logistic regression, the dichotomized utilization variables were used as the dependent variables. Finally, Spearman partial correlations were performed to illustrate the relationships between the demographic and risk variables and the caregiver reported severity variables. Analyses were performed using SAS software. All tests were 2-sided and were considered significant at = 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
1 hospitalization for breathing problems between birth and study enrollment (range: 0–7). Fifty-nine percent had 2 ED visits during that time (range: 0–11). Parent ratings of the children’s functional illness severity averaged 1.57 (±0.9), and the average quality of life score was 5.67 (±1.3). Foreign-born Hispanic families had taken their infants to the ED significantly more than other groups. Although there was a significant correlation between ED visits and infant age (rs = 0.15, P = .05), this did not account for ethnic group differences because children in the different groups were the same age at study entry (P = .38). Subsequent analyses adjusted for ethnic group membership and infant age.
|
2 analyses showed that children receiving controller medications were more likely to have been hospitalized (P < .01), but there was no relationship between receiving controller medications and having 2 or more ED visits (P = .30). Corticosteroid bursts were associated with hospitalizations (P < .001) and with ED visits (P < .001).
Correlates of Health Care Utilization
As shown in Table 3, Spearman correlations were used to assess associations between hospitalizations (number/age in months at study entry) and infant and family characteristics, perinatal variables, biological exposures occurring between birth and study entry, and environmental and psychosocial characteristics assessed at study entry. Higher caregiver cognitive scores were associated with fewer hospitalizations. As expected, presumed or confirmed RSV coded from the medical record was significantly associated with hospitalizations. In addition, elevated IgE levels and evidence of cockroach allergen in the home were associated with hospitalization. None of these relationships were affected by the adjustment for ethnic group. However, a significant relationship between current anxiety and infant hospitalizations since birth was reduced to a trend after adjusting for ethnic group membership.
|
Multiple logistic regression was used to determine if allergy variables were associated with hospitalization after RSV was taken into account. Table 4 shows that after adjusting for ethnic group and infant age, RSV, cockroach allergen, and elevated IgE entered the model, indicating that each of the variables was significantly and independently related to 1 or more child hospitalizations. Caregiver cognitive score did not enter the model. Examination of the variables for colinearity showed a significant relationship between caregiver cognitive scores and presence of cockroach in the home (adjusted r = –0.16, P < .04).
|
Caregiver Morbidity Reports
The Spearman correlations in Table 5, adjusted for ethnic group, show univariate relationships between the medical utilization, demographic, biological, environmental exposure, and psychosocial variables in relation to caregiver reports of the infants’ functional severity of wheezing illness and wheezing-related quality of life (CQoL). There was no relationship between the caregivers’ reports of illness severity and the medical record coding of hospitalizations. However, there was a significant relationship between illness severity scores and number of ED visits since the child’s birth. Children whose biological mothers had asthma were reported to have significantly greater illness severity. In addition, parent report of greater illness severity was associated with low maternal cognitive score. Caregivers’ reports of their children’s illness severity were significantly related to higher anxiety levels, as well as to higher numbers of stressful events reported having occurred during the past year. In practical terms, each variable that was significantly related to parent report of illness severity influenced the maternal report 
0.4 in the indicated direction on a severity scale ranging from 0 to 4 (mean = 1.57, SD = 0.9). Caregivers’ reports of their quality of life with regard to their children’s breathing problems were associated only with their concurrent level of anxiety; higher quality of life was highly significantly related to low maternal anxiety.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Ethnic group was an important variable associated with morbidity patterns, whether caregiver reported or medical record derived. A majority of the families in the study were Hispanic, originating primarily from Mexico, but we found important differences when we separated the Hispanic group according to the caregiver’s country of birth.19 Although the US-born Hispanic families had morbidity patterns similar to the whites and blacks, the foreign-born Hispanic families had significantly higher numbers of ED visits. At the same time, they reported less illness severity in their children. The severity reports provided by these Spanish-speaking caregivers raise questions about the manner in which language and culture might influence caregiver reports. Health knowledge, such as perception of symptoms, or health beliefs, such as comfort in reporting symptoms, differ across cultural and language groups. Influences on styles of symptom reporting may be largely unrelated to factors that influence use of the health care system.
It is well-known that for many inner-city minority children, the primary place for asthma care is the ED.17,37 Numerous factors have been suggested to account for this qualitatively different pattern of utilization of the health care system, including inadequate access to health care, financial concerns, and psychosocial characteristics. Increased utilization of the ED has been related to health beliefs and poor illness knowledge among Hispanics,38 Native Americans,39 and blacks.40 It is unclear why the foreign-born Hispanic caregivers used the ED significantly more than US-born Hispanic families, blacks, or whites. It seems likely that, as the immigrant group in this study, these caregivers lacked knowledge of optimal ways to access the US medical system, or financial or language barriers prevented them from accessing alternative sources of care. However, Zambrana et al41 studied the use of emergency pediatric services among immigrant, primarily Spanish-speaking Latin mothers, and found that Mexican immigrants were 1.5 times more likely than Central American immigrants to use the ED as their usual source of care for their children’s general medical needs. We are continuing to examine the barriers to accessing medical care for the families in the CAPS study (J. Hollander, M. Price, P. Barton, and M. Klinnert, submitted for publication).
A number of biological variables were associated with the medical record health care utilization indexes of morbidity. As expected, RSV infections were associated with hospitalizations. Lower respiratory infection, in particular RSV, is one of the primary variables associated with wheezing episodes in infancy.10,42 The data also suggested a role for allergy as an additional and independent contributor to morbidity of the infants’ wheezing illnesses, as shown by the association of elevated IgE with hospitalizations and ED visits. Furthermore, the presence of cockroach allergen in the homes was a significant, independent predictor for hospitalizations. This is consistent with the very important role that has been found for cockroach allergen in infant wheezing42,43 as well as in childhood asthma.44 For childhood asthma, the combination of sensitization to cockroach allergen and presence of allergen in the home has been associated with high morbidity among inner-city children.44 Low numbers of specific sensitivities were expected in the study reported here, in which the children were an average of 16 months old. Of the 30 children in this study with at least 1 positive skin test, only 1 child showed sensitivity to cockroach. This is consistent with findings of others,42 who found that even very low levels of cockroach allergen in homes were associated with incident asthma in young children, although sensitization status for those children was not known.
Increased use of the ED was associated not with biological or environmental variables, but with parenting singly and with the caregiver smoking. The latter is consistent with Duff’s findings that infants brought to the ED for wheezing had high cotinine levels.10 Others have found that cotinine levels in young children are highly correlated with the smoking habits of their mothers.45,46 In this study, cotinine levels were not associated with ED visits or hospitalizations, probably because the medical events were not closely related in time with the children’s urine collection at study entry. Parenting singly, also associated with more ED visits, was significantly more common among the smoking mothers. Sixty-five percent of the single caregivers smoked compared with 35% of those with partners, suggesting that increased smoke exposure may be one reason that children of single caregivers had more ED visits. However, analyses showed that both parenting singly and caregiver smokes contributed independently to ED visits. Rand,17 who found that parenting alone was associated with ED visits among low-income black school children, suggested that the lack of social support existing in single parent homes may lead to ED visits through inadequate instrumental help in accomplishing asthma management tasks or obtaining preventive health care.
Several psychosocial variables were significantly associated with both increased hospitalizations and ED visits. Lower cognitive scores were associated with more ED visits, but this effect disappeared after adjustment for ethnic group, because the adjustment took into account the fact that the foreign-born Hispanic families as a group had more ED visits and they also tested lower on the cognitive screen. In contrast, lower caregiver cognitive scores were associated with more hospitalizations both before and after adjustment for ethnic group. There was no such relationship between caregiver education levels and the utilization variables. This is likely because there was little variability in the education levels, whereas the cognitive screening yielded a wide range of scores. Education functions as a proxy for cognitive ability only in a culture with access to a variety of educational opportunities. Cognitive scores are related in large part to exposure to domains of knowledge, and lower scores are likely related to less knowledge of illness management for wheezing infants. Lower asthma knowledge was significantly associated with increased hospitalizations for childhood asthma in a Hispanic sample in the Southwest.38 Lower cognitive scores also may be indicative of poorer problem-solving skills and poorer judgment, which can be expected to lead to greater health care utilization.47 Interestingly, the multivariate analysis showed that cognitive scores did not contribute to more hospitalizations after cockroach levels were taken into account, presumably attributable to the fact that high cockroach scores and low cognitive scores were correlated.
The morbidity variables based on caregiver report had a somewhat different pattern of determinants in comparison with the health care utilization variables. Maternal reports of functional severity were unrelated to biological factors such as child atopy or environmental exposures (data not shown). However, close examination of the relationships showed some consistencies in the data. Caregiver reports of illness severity were unrelated to frequency of hospitalizations, but there was a significant relationship with the number of objectively coded ED visits. This may reflect the fact that many of the hospitalizations were precipitated by acute respiratory illnesses with wheezing (eg, RSV) in children who did not have the chronic breathing problems captured with the illness severity measure. In contrast, children taken more often to the ED were reported by caregivers to have more frequent and chronic wheezing problems.
Analyses showed that mothers with asthma reported greater illness severity in their children. This is suggestive of biological maternal influence such as genetic transmission or prenatal conditions.11 However, the increased severity was apparent only in the caregiver report and not in the health care utilization data. This may indicate that maternal asthma influenced report of infant illness through other avenues. For example, mothers with asthma may perceive wheezing illness in their children more readily than those who are inexperienced with asthma and may report greater frequency. Or, if the infants of mothers with asthma had more severe illness, these caregivers who were experienced with asthma management may have been less likely to take their wheezing infants to the ED, and may even have had more skills with early symptom detection and treatment to prevent hospitalization.
Caregiver anxiety and stress showed strong associations with caregiver-reported illness severity. Interestingly, caregiver anxiety was not associated with either ED visits or hospitalizations; in fact, there was a nonsignificant trend toward lower anxiety being associated with hospitalizations. In contrast, caregivers reporting high anxiety for themselves also reported greater illness severity in their children. Their increased anxiety may have resulted from their children’s illness severity, or at least from their perception of the children’s illness severity. It is of interest that in this study there was no relationship between maternal anxiety and objectively coded ED use, hospitalizations, or medication level. On the other hand, because the reports are from the same informant for the same time period, it is also possible that the maternal anxiety contributed to the report of increased severity. Mothers who are anxious may be more vigilant and perceptive of breathing problems in their children, thus reporting higher frequencies for the items on the severity scale. It is also possible that their anxiety affects their response to the scale itself, through over-reporting symptoms and inflation of the score.
The final measure of morbidity, CqoL, was associated most strongly with caregiver anxiety. A strong relationship was expected, because the quality of life scale is heavily focused on emotional responses to children’s illness.22 However, it was surprising that the caregivers’ reports of quality of life were completely unassociated with their infants’ hospitalizations or ED visits.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We thank Dr James Murphy and Dr Harold Nelson. We are grateful to the health care providers and clinics and to the families who participated in CAPS.
|
FOOTNOTES |
|---|
Reprint requests to (M.K.) National Jewish Medical and Research Center, 1400 Jackson St, Denver, CO 80206. E-mail: klinnertm{at}njc.org
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. D. Klinnert, A. H. Liu, M. R. Pearson, S. Tong, M. Strand, A. Luckow, and J. L. Robinson Outcome of a Randomized Multifaceted Intervention With Low-Income Families of Wheezing Infants Arch Pediatr Adolesc Med, August 1, 2007; 161(8): 783 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Klinnert, A. H. Liu, M. R. Pearson, M. C. Ellison, N. Budhiraja, and J. L. Robinson Short-term Impact of a Randomized Multifaceted Intervention for Wheezing Infants in Low-Income Families Arch Pediatr Adolesc Med, January 1, 2005; 159(1): 75 - 82. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2003 American Academy of Pediatrics. All rights reserved. |
||
|
||
