Nocturnal Asthma in Children Affects School Attendance, School Performance, and Parents' Work Attendance | Asthma | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Sign In
September 2000

Nocturnal Asthma in Children Affects School Attendance, School Performance, and Parents' Work Attendance

Author Affiliations

From the Divisions of Pulmonary and Critical Care Medicine (Dr Diette) and General Medicine (Dr Wu), and the Department of Pediatrics (Dr Nguyen), School of Medicine, and the Departments of Epidemiology (Drs Diette and Wu) and Health Policy and Management (Ms Skinner and Dr Wu), School of Hygiene and Public Health, Johns Hopkins University, Baltimore, Md; and Merck & Co Inc, West Point, Pa (Drs Markson and Algatt-Bergstrom).

Arch Pediatr Adolesc Med. 2000;154(9):923-928. doi:10.1001/archpedi.154.9.923

Context  Asthma symptoms that occur at night may signal worse asthma control, but the nighttime occurrence may have additional clinical significance. To date, however, there have been few studies of the impact of nocturnal awakening from asthma on children with the disease, including problems with daytime functioning.

Objective  To determine if school absenteeism and school performance in children and work absenteeism in their parents are associated with nocturnal awakenings from asthma.

Design  Cross-sectional survey during the winter of 1997 through 1998.

Setting  Three managed care organizations in the United States.

Participants  Parents of 438 children with asthma, aged 5 to 17 years, who were enrolled in managed care organizations.

Intervention  None.

Main Outcome Measures  Parent's reports of number of days their child missed school and parent missed work and how often the child's education suffered because of asthma in the past 4 weeks.

Results  Overall, more than 40% of children had nocturnal awakenings from asthma in the past 4 weeks. Multivariate analyses were performed that adjusted for child age, race, overall symptom severity, and use of reliever medications. Compared with children who did not awaken from asthma, there were greater odds of missed school days in children who awakened 1 to 3 nights (odds ratio [OR], 3.6; 95% confidence interval [CI], 2.1-6.2), 4 to 7 nights (OR, 4.4; 95% CI, 2.0-10.0), and more than 7 nights (OR, 14.7; 95% CI, 5.9-37.0). Similarly, there were greater odds of education suffering in children who awakened 1 to 3 nights (OR, 2.3; 95% CI, 1.4-3.7), 4 to 7 nights (OR, 2.1; 95% CI, 0.9-4.6), and more than 7 nights (OR, 2.3; 95% CI, 1.0-5.4), and parents missing work in children who awakened 1 to 3 nights (OR, 4.0; 95% CI, 2.2-7.1), 4 to 7 nights (OR, 6.5; 95% CI, 2.7-16), and more than 7 nights (OR, 3.2; 95% CI, 1.3-7.9). Greater overall symptom severity and high use of reliever mediation were also associated with missed school, education suffering, and parent absenteeism.

Conclusions  Nighttime awakenings in children with asthma may affect school attendance and performance, as well as work attendance by parents. Nighttime symptoms have independent prognostic value, even when overall asthma symptom severity is accounted for. By addressing whether there are nighttime awakenings in children with asthma, clinicians may be able to tailor the therapeutic regimen to counter these symptoms.

ASTHMA SYMPTOMS that occur at night are interpreted by clinicians as asthma that is not well controlled. However, if nighttime awakenings were associated with consequences to the patient that were independent of overall symptom severity, nocturnal asthma symptoms would take on a broader significance for clinical treatment of pediatric asthma. An illness that disrupts the quantity or quality of sleep might be expected to affect daytime functioning.1,2 For children, this disruption may include school performance, and for parents of these children, it may include work performance.

In this study, we examined survey data collected from parents of children with asthma to learn whether nocturnal asthma symptoms are simply a marker of overall symptom severity or whether they convey independent prognostic information. Specifically, the purpose of this study was to determine whether nocturnal awakenings from asthma, independent of overall symptom severity, are associated with children's school attendance and performance, and with their parents' attendance at work.

Materials and methods

Study design

This study employed a cross-sectional survey of parents of children with asthma who were enrolled in 3 large managed care organizations (MCOs) in the United States during the winter of 1997 through 1998. A major goal of the study was to examine the impact of asthma on children and their families. The study was reviewed and approved by the Committee on Human Research of the Johns Hopkins University School of Hygiene and Public Health, Baltimore, Md.

Study sites

Three MCOs in the Northeast and Midwest United States volunteered to participate in the project. Study participants were selected from the pool of enrollees in each MCO using claims data and centralized pharmacy data. The sampling specifications were to identify at least 300 children with asthma in each MCO aged 5 to 17 years. To assure a sufficient number of older children, the sampling frame was stratified by age (5-10 years and 11-17 years), with at least 150 children to be selected for each stratum.

There was a 2-step process for identifying patients with asthma. In the first step, MCOs used administrative and/or pharmacy data to identify potential asthma patients. Using administrative data, children were identified who had 2 or more visits (outpatient or inpatient) for asthma in the last 12 months (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM],3 code 493.x). Pharmacy data were used to identify patients with 1 or more dispensations of a medication commonly prescribed for asthma. The pharmacy dispensation records were searched by Food and Drug Administration National Drug Classification codes for medications that included corticosteroids, sympathomimetics, anticholinergics, xanthines, and cromones. In the second step, responses from the survey were used to exclude children whose parents indicated that their children did not have asthma.

The survey design was based on a mailed survey to parents of sampled children, with follow-up efforts by telephone and mail. An advance letter, signed by an MCO official, explained the study to parents and served as an informed consent statement (voluntary nature of participation, procedures to protect confidentiality of information provided, and absence of repercussions in dealings with the MCO and/or medical care provider whether or not they decided to participate). The mailed survey was accompanied by a cover letter reiterating the information contained in the advance letter and by a stamped self-addressed envelope. Respondents were considered ineligible if the parent indicated that the child did not have asthma, or was not currently enrolled in the MCO.

Survey measures

The survey included questions that assessed general health status, asthma symptoms, impact of asthma on physical and mental health, family and patient role functioning, health service utilization, ratings of the quality of care, asthma-related education and participation in care, self-management abilities, use of treatments and medications, and parent and child demographics. Sources of survey items included the Pediatric and Adolescent Asthma Therapy Assessment Questionnaire,4 the New England Medical Center Asthma Outcomes Monitoring System child asthma questionnaire,5 the Managed Health Care Association Outcomes Management System Asthma Study questionnaire,6 the Maryland Medicaid Recipients Study survey,7 and the Child Health and Illness Profile.8

Outcome measures

Outcomes assessed for the prior 4-week period included parent reports of the following: (1) days of missed school because of asthma (none, 1-3, 4-7, and >7); (2) how often the child's education suffered because of asthma (all, most, some, and a little or none of the time); and (3) the number of days the child's asthma kept the parent from work or other usual activities (measured as a continuous variable). For analysis, parent days of missed work were grouped into ordinal categories (none, 1-3, and ≥4).

Explanatory variables

Survey items inquired about child demographics (age, sex, and race) and parent characteristics, including education and employment status (full-time, part-time, or not employed). Questions about asthma symptoms assessed the frequency of specific symptoms over the prior 4-week period. Nocturnal symptoms were assessed by the number of nights the child was awakened with wheezing or difficulty in breathing (none, 1-3, 4-7, and >7). Other question items assessed the frequency of wheeze, chest tightness, cough, shortness of breath, and flares and the degree of symptoms between attacks. Flares were defined as "difficulty breathing that interfered with normal activities and that may have been accompanied by coughing, wheezing, shortness of breath or other symptoms." Parents were also asked about treatments the child used, including medication types and their frequency.


Basic descriptive statistics were generated, including proportions and means. For bivariate analyses, outcomes were dichotomized (none vs any) and the percentages of children with each outcome were examined by the number of nights awakened, with significance determined by the χ2 test for trend. As an initial step to examine whether the relationship of nocturnal awakening to functional consequences was independent of other indicators of symptom severity, we examined the relationship of nocturnal awakening to the selected outcomes in the stratum of children with severe persistent symptoms9 (defined as symptoms of wheeze, chest tightness, cough, shortness of breath >1 per day, or ≥4 flares per week).

Multivariate analyses were conducted as a 2-stage process. Independent variables were selected by clinical judgment and the results of a bivariate analysis that examined the relationship of potential confounders to each of the 3 outcome variables (missed school days, education suffered, and missed work by parent because of asthma). Statistical tests that were employed to assess significance included χ2 for nominal and dichotomous factors and χ2 test for trend for ordinal factors. Factors were selected for the multivariate models if P<.05 for at least 1 of the outcome measures. Candidate independent variables included age, sex, race, Hispanic origin, parent education, parent employment, use of medications associated with asthma severity (frequent use of inhaled β-agonists), and symptom severity. Frequent use of reliever medications was defined as use more than 4 times per day.10 A measure of symptom severity was constructed to approximate the scheme outlined in the 1997 National Asthma Education and Prevention Program guidelines.9 An approximation was used as there is no precise measurement tool included in the guidelines that can be applied to research. We categorized patient symptoms ordinally (mild intermittent or mild, moderate, or severe persistent) based on ordinal responses to questions about the frequency of wheeze, chest tightness, shortness of breath, cough, and flares, as well as symptom status on days in between flares or attacks. In bivariate and multivariate analyses, severity was characterized ordinally (mild intermittent [1] to severe persistent [4]). To determine whether the assumption of independence of potential severity-related variables was appropriate, we performed a correlation analysis (Spearman rank correlation coefficient)11 of frequency of nocturnal awakening, frequency of reliever medication use, and overall symptom frequency. Correlation was low to moderate (overall symptoms vs reliever use, 0.19; nocturnal awakening vs reliever use, 0.34; and nocturnal awakening vs overall symptoms, −0.38), so all of these factors were entered into the models. Multivariate analyses were conducted using ordinal logistic regression, with P<.05 to denote significance. Ordinal logistic regression, an adaptation of logistic regression, can be used to estimate relationships between an ordinal dependent variable and a set of independent variables.12 For the multivariate models reported here, the c statistic ranged from 0.73 to 0.81. Statistical analyses were performed with the SAS statistical software package, version 6.12 (SAS Institute, Cary, NC).


Of 990 patients identified by the 3 MCOs, 894 (90%) were eligible. Reasons for ineligibility included no asthma (n = 89), not enrolled (n = 5), and a language barrier that prevented completion of the survey (n = 2). Of eligible patients, 438 (49%) completed the survey. There was limited information available about nonrespondents at 2 of the 3 MCOs. Nonresponse was greater among parents of children who were older (aged 11-17 years) and female, and in one of the MCOs, nonresponse was greater for parents enrolled in point-of-service plans rather than MCOs.

Of the 438 children, 74% were white and 19% black; 57% of the children were male (Table 1). More than 40% of the children had been awakened at least 1 night with asthma symptoms in the 4 weeks before the survey and 66% had experienced exercise-related symptoms. During the same period, parents reported that 35% of children missed at least 1 day of school because of asthma symptoms and 36% reported that school suffered because of asthma.

Table 1. 
Characteristics of Children and Their Parents
Characteristics of Children and Their Parents

There was a significant trend with more nights awakened in the past 4 weeks and the likelihood of missing school, education suffering, and parents missing work or other usual activities (P<.001 by χ2 test for trend) (Table 2). While most children (77%) with more than 7 nights of awakening missed at least 1 day of school, more than one half of children missed school when awakened as few as 1 to 3 nights (P<.001). Similarly, 56% of parents of children with more than 7 nights awakened missed work/usual activities, as did 44% of parents whose child awakened 1 to 3 nights (P<.001). In the subset of parents who were employed full-time (162 of 438 responding parents), only 10% missed a day of work with children who did not awaken from asthma, but 39% and 60% missed work whose children awakened 1 to 3 and 4 to 7 nights, respectively (P<.001, by χ2 test).

Table 2. 
Relationship of Nocturnal Awakening to Children's Daytime Functioning in the Past 4 Weeks
Relationship of Nocturnal Awakening to Children's Daytime Functioning in the Past 4 Weeks

Stratification of results by overall symptom severity suggested that nocturnal awakening may be an independent determinant of the selected study outcomes. There was a gradient between the frequency of nocturnal awakening and functional consequences even among children with severe symptoms (Table 2). Among children with severe symptoms, 58% missed school because of asthma when awakened 1 to 3 nights; but if there were no nocturnal awakenings, only 20% missed school (P<.001). All 3 outcomes were associated with the number of nights awakened, even among patients with severe symptoms.

Table 3 shows results of additional bivariate analyses. As a reflection of either disease severity or lack of asthma control, frequent use of reliever medication, and persistent symptoms were significantly related to all 3 outcomes. Parents of younger rather than older children were more likely to interrupt work or other activities (35% vs 24%) (P<.05), and black children were more likely to miss school than children of other races. Parent education and employment status were not significantly related to any of the 3 outcomes (data not shown).

Table 3. 
Bivariate Analysis of the Relationship of Other Child Characteristics to Daytime Functioning
Bivariate Analysis of the Relationship of Other Child Characteristics to Daytime Functioning

The main observations from the bivariate analyses were not significantly affected by multivariate adjustments (Table 4). Nighttime awakenings, overall symptom severity, and frequent use of reliever medications were all associated with functional consequences, even after adjustment for child demographics. The number of school days missed was strongly related to increasing numbers of nights awakened, with odds ratios (ORs) of 3.65 (95% confidence interval [CI], 2.14-6.20), 4.44 (95% CI, 1.95-10.1), and 14.7 (95% CI, 5.89-36.6), for 1 to 3, 4 to 7, and more than 7 nights, respectively. Waking up at least 1 night was significantly associated with education suffering and parents missing work, although there was not a dose-response relationship seen. In these multivariate analyses, black children were more likely (OR, 2.31 [95% CI, 1.32-4.04]) to miss school than white children and parents of older children were slightly less likely to miss work than parents of younger children (OR, 0.88 [95% CI, 0.82-0.94]).

Table 4. 
Multivariate Models From Ordinal Logistic Regression* of the Relationship of Characteristics of Children to Daytime Functioning†
Multivariate Models From Ordinal Logistic Regression* of the Relationship of Characteristics of Children to Daytime Functioning†


The major findings in our study are that the presence of nocturnal asthma symptoms in children is associated with decreased school attendance and performance, as well as with missed work in their parents. Nocturnal asthma in children may be more disruptive to children and their parents than has been previously recognized. Particularly striking was the observation that approximately one half of children with even a small number of nights awakened in a 4-week period (1 to 3 nights) missed school and had education suffer, and nearly one half of their parents missed 1 or more days of work.

There have been few previous studies of the effect of nocturnal asthma on daytime performance and morbidity, especially in children. Recognizing this gap in current knowledge, the National Heart, Lung, and Blood Institute's recent workshop on the chronobiology of asthma recommended future research toward understanding the daytime sequelae of nocturnal asthma, including school performance in children. In an analysis of cross-sectional survey data from the US National Health Interview Survey, Fowler et al13 showed that children with asthma had a higher prevalence of grade failure, learning disabilities, and absenteeism. However, this study did not assess severity of asthma nor whether asthma had awakened the children at night. Fitzpatrick et al2 studied 12 patients with nocturnal asthma and compared them with 12 normal controls, by examining sleep diaries, polysomnograms, and tests of cognitive function. They found that the asthma patients had poorer subjective sleep quality, more time awake during the night, and lower scores on tests of daytime cognitive performance. A more recent study by Stores et al14 compared subjective ratings of sleep quality and daytime sleepiness, and tests of cognitive function, mood, and behavior in children with nocturnal asthma with children without asthma. Children with nocturnal asthma had greater disruption of sleep, greater daytime sleepiness, and demonstrated lower cognitive function on a test of memory recall. After changes in their medication regimens were instituted, the children with asthma had improvements in sleep disruption, daytime sleepiness, and cognitive function.

One mechanism for how nocturnal asthma could lead to problems with daytime function is reduction in the quantity and quality of sleep. In the study of subjective and objective disturbance of sleep by Stores et al,14 children with asthma averaged 23 minutes less sleep per night (not significant) and had significantly greater numbers of awakenings compared with nonconcurrent controls. Sleep efficiency (the percentage of time in bed that one is asleep), an indicator of sleep quality, was lower in asthmatics than controls in the study by Fitzpatrick et al.2 A connection between asthma and the pathologic condition of sleep-disordered breathing was shown in a recent study by Redline et al.15 In that study of 399 children and adolescents, the history of physician-diagnosed asthma and self-reported wheeze were independently associated with sleep-disordered breathing, as assessed by polysomnogram.

Findings in the present study (that overuse of reliever medication was associated with functional consequences even after accounting for symptom severity) suggest that disease severity or lack of disease control were reflected in this measure. Previous studies have suggested that higher use of inhaled β-agonist medications may be an indicator of poorly controlled asthma.10,16,17 The National Asthma Education and Prevention Program guidelines suggest that frequent use of reliever medications is a sign of greater symptom severity, and recommend that daily use of this type of medication should prompt an increase in long-term control therapy.

The current study findings may have relevance for asthma classification. Currently, several severity classifications of asthma combine daytime and nighttime symptoms. To the extent that symptom classification should relate to prognosis, our results suggest that, for certain outcomes, it may be useful to separate overall symptom frequency from the frequency of symptoms at night, and to consider if the presence of nocturnal awakenings is a reason to classify patients as more severe.

It should be noted that data in this study are cross-sectional, so inferences of causality cannot be made. For example, while it is possible that nighttime symptoms cause children to miss school, this study design does not exclude the possibility that staying home from school leads to contact with asthma triggers that in turn increase nighttime symptoms. Our measure of school performance was based on the parents' perceptions that school had suffered. Although parents may have believed that school performance suffered (eg, because they knew that the child missed school or was sleepy from staying up at night), this perception was not validated with objective measures such as grades, test scores, or teacher reports. While this study has shown an association between nocturnal awakening during the winter season, it is possible that the impact of nighttime awakening differs in other seasons. This type of study may bear repeating in the summer, for example, when children are out of school and may have fewer daytime obligations. Our patient sample is unlikely to be representative of all children cared for by the study MCOs with the sample biased by parents more willing to volunteer for the study. Parents whose children did not wake up from asthma may have been less likely to participate in the survey, which may have affected our estimates of the associations of nighttime awakening with daytime consequences. All data were from parent report and are subject to limitations of recall and possible recall bias. It is possible, for example, that those with school and work limitations may have better recall of nighttime sleep disturbance, in which case we may have overestimated the impact of nighttime symptoms.


To our knowledge, this study is the first to demonstrate the potential impact of nocturnal asthma symptoms on school attendance and performance in children and work attendance in their parents. We have shown that not only is assessment of overall asthma symptom severity related to these functional consequences, but also that report of nighttime awakening has independent explanatory value. Clinicians might consider assessing whether asthma is awakening children at night, and target the reduction of nocturnal symptoms as a goal of therapy. When children are awakened from asthma, treatment strategies can be employed to target the reduction of these nighttime symptoms.18,19 Such interventions that may be effective include optimization of medication regimens directed at long-term control and environmental controls in the home. Clinicians should also ask about frequent use of reliever medications, as such use can be an indication that the therapeutic regimen may need adjustment or intensification.

Accepted for publication March 29, 2000.

Support for this project was provided by a grant from Merck & Co Inc, West Point, Pa (Drs Diette and Wu and Ms Skinner).

We thank Becky Clark, BA, for her help with statistical programming for the data analyses, and Philip Smith, MD, and Jerry Krishnan, MD, for their comments on an earlier version of this manuscript.

Corresponding author: Gregory B. Diette, MD, MHS, Room 301, 1830 E Monument St, Baltimore, MD 21205 (e-mail:

Vir  RBhagat  RShah  A Sleep disturbances in clinically stable young asthmatic adults.  Ann Allergy Asthma Immunol. 1997;79251- 255Google ScholarCrossref
Fitzpatrick  MFEngleman  HWhyte  KFDeary  IJShapiro  CMDouglas  NJ Morbidity in nocturnal asthma: sleep quality and daytime cognitive performance.  Thorax. 1991;46569- 573Google ScholarCrossref
Not Available, International Classification of Diseases, Ninth Revision, Clinical Modification.  Washington, DC Public Health Service, US Dept of Health and Human Services1988;
Wu  AWAlgatt-Bergstrom  PDiette  GB  et al.  Development of the pediatric control score [abstract].  Am J Respir Crit Care Med. 1999;159A911Google ScholarCrossref
Ware  JEBayliss  MS Outcomes monitoring in asthma: measuring generic and asthma-specific functional health and well-being, clinical and cost outcomes. Presented at the 2nd Annual Disease Management Congress June 18, 1997 New York, NY
Steinwachs  DMWu  AWSkinner  EAYoung  Y Outcomes management system: asthma final report. Report to Managed Health Care Association, Outcomes Management System Project1996; Washington, DC.
Rubin  HRJenckes  MStuart  M Maryland Mediciad recipients ratings of prenatal and pediatric care by HMOs and fee-for-service providers. Presented at the 12th Annual Meeting of the Association for Health Services Research June 5, 1995 Chicago, Ill
Starfield  BBergner  MEnsminger  M  et al.  Adolescent health status measurement: development of the Child Health and Illness Profile.  Pediatrics. 1993;91430- 435Google Scholar
National Heart Lung and Blood Institute, Guidelines for the Diagnosis and Management of Asthma: Expert Panel Report No. 2.  Bethesda, Md National Institutes of Health1997;NIH publication 97-4051.
Diette  GBWu  AWSkinner  EA  et al.  Treatment patterns among adult asthmatics: factors associated with overuse of inhaled β-agonists and underuse of inhaled corticosteroids.  Arch Intern Med. 1999;1592697- 2704Google ScholarCrossref
Rosner  B Nonparametric methods.  Fundamentals of Biostatistics Belmont, Calif Duxbury Press1995;551- 584Google Scholar
Greenland  S An application of logistic models to the analysis of ordinal response.  Biometric J. 1985;27189- 197Google ScholarCrossref
Fowler  MGDavenport  MGGarg  R School functioning of US children with asthma.  Pediatrics. 1992;90939- 944Google Scholar
Stores  GEllis  AJWiggs  LCrawford  CThomson  A Sleep and psychological disturbance in nocturnal asthma.  Arch Dis Child. 1998;78443- 449Google ScholarCrossref
Redline  STishler  PSchluchter  MAylor  JClark  KGraham  G Risk factors for sleep-disordered breathing in children: association with obesity, race, and respiratory problems.  Am J Respir Crit Care Med. 1999;1591527- 1532Google ScholarCrossref
Spitzer  WOSuissa  SErnest  P  et al.  The use of β-agonists and the risk of death and near death from asthma.  N Engl J Med. 1992;326501- 506Google ScholarCrossref
Speizer  FEDoll  RHeaf  PStreng  LB Investigation into use of drugs preceding death from asthma.  BMJ. 1968;1339- 343Google ScholarCrossref
Martin  RJKraft  M Nocturnal asthma: therapeutic considerations.  Clin Immunother. 1996;6443- 453Google ScholarCrossref
Desjardin  JABallard  RD Nocturnal asthma: mechanisms and current therapy.  Drug Ther. 1993;2329- 38Google Scholar