| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Academic Division of Clinical Psychology, University of Manchester (R.C., L.G.), UK; and Institute of Psychiatry (R.G.), King’s College London, UK.
Address correspondence and reprint requests to Dr. Rachel Calam, Academic Division of Clinical Psychology, University of Manchester, Second Floor, Education and Research Centre, Wythenshawe Hospital, Manchester M23 9LT, UK. E-mail:Rachel.Calam{at}man.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: Survey design with random selection based on national computer records of 10,438 children aged 5 to 15 from 12,529 eligible families (83%). Strengths and Difficulties Questionnaire (SDQ) was completed by parents, teachers, and 11- to 17-year-olds, providing scores for total difficulties, emotional symptoms, conduct problems, hyperactivity, peer problems, social behavior, and total impact. The Development and Well-Being Assessment (DAWBA) generated ICD-10 diagnoses. Parents reported physical disorders and disabilities, other child mental health risk factors, and rated child’s general level of health.
Results: Children with organic conditions other than asthma were excluded, leaving 9,834 children, 49.9% male. Preliminary comparisons indicated higher parent and teacher rated scores for children with asthma. New variables combined asthma and general health to produce four groups. Logistic regressions were conducted with children without asthma and in good health as the reference group. Children with asthma in good health showed greater parent-rated emotional problems, but otherwise, few elevated odds ratios. Children without asthma in poor health were at significantly greater risk of disorder, as were children with asthma in poor health. Findings on ratings of hyperactivity are discussed.
Conclusion: Findings that children with asthma have elevated psychological difficulties may result from poor health rather than asthma itself.
Key Words: asthma • adaptation–psychological • child behavior
Abbreviations: SDQ = Strengths and Difficulties Questionnaire; DAWBA = Development and Well-Being Assessment.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
It is possible that it is not asthma per se that leads to elevated ratings of psychological difficulties, but that children who are unwell or whose asthma is poorly controlled are at particular risk of being reported as having psychological difficulties. For children with chronic illnesses, poor functional status and school absences are associated with poorer psychological adjustment (10). It may therefore be the case that the extent to which the child is unwell with asthma has a significant impact on psychological adjustment. Recently, for example, a cohort study (11) showed that parents of 3-year-olds showing recurrent attacks of severe wheeze and shortness of breath reported significantly higher levels of child behavioral problems.
In their meta-analysis of studies of behavioral adjustment in children with asthma, McQuaid et al. (1) noted a number of methodological difficulties in the field. The samples of convenience and clinic samples making up the majority of studies are not as methodologically sound as large random samples. Furthermore, where comparison groups are used, these are usually healthy controls, so that the effects of poor health per se are not assessed. Studies do not always control for socioeconomic status or other important demographic factors. The majority of studies rely on parent reports, and few studies use standardized teacher reports. Self-report of adjustment by children is rarely included.
Analysis of data from the 1999 UK nationwide child mental health survey of children aged 5 to 15 (12) allowed us to examine relationships between the presence or absence of asthma and reported psychological disturbance within a large population sample, while controlling for other factors. The data set allowed us to investigate the impact of the children’s overall health status on parental ratings, teacher ratings, self-report by the child, and psychiatric diagnosis. The measure of asthma, that is, parental report of presence or absence, was basic, but parents also gave a rating of general health of the child. It has already been established that a wide range of variables has an impact on the child’s adjustment. Prior analysis of the contribution of a range of variables, including child, family, school, and neighborhood factors within the sample (13) enabled us to anticipate a wide range of variables that might potentially have a bearing on any statistical relationships found between asthma and psychological adjustment, and these were controlled for in analysis. We therefore controlled for a range of factors, including gender, socioeconomic factors, and parental mental health, enabling us to assess the independent effect of asthma and general health on risk for behavioral and emotional disturbance and psychiatric diagnosis for the child. The survey data enabled us to test the hypothesis that children with asthma are at increased risk of hyperactivity on a large population sample.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Measures
Psychopathology
The Strengths and Difficulties Questionnaire (SDQ) is a brief behavioral questionnaire with good psychometric properties (14) generating dimensional scores for total difficulties, emotional symptoms, conduct problems, hyperactivity, peer problems, prosocial behavior, and total impact. It can be completed by parents, teachers, and 11- to 17-year-olds.
The Development and Well-Being Assessment (DAWBA) consists of a structured interview and questionnaires administered to parents, teachers, and young people by lay interviewers who record verbatim accounts of reported problems (15). Child and adolescent psychiatrists then reviewed both verbatim accounts and answers to structured questions about symptoms and their impact (resultant distress and social impairment) and then assigned diagnoses according to ICD-10 criteria (16).
Specific Physical Disorders and Disabilities
Parents completed 3 checklists, each listing 12 physical and psychological difficulties. They were asked: "Please can you tell me if [Child] has ... ". Listed conditions included "Asthma," "A heart problem," "Epilepsy," "Diabetes," and "Cerebral palsy." Each item could be answered "Yes" or "No."
Data on other possible risk factors for child mental health problems were collected from parents. These included a rating of general health of the child. The parent was asked to rate their child’s health on a 5-point scale, with the question "How is (CHILD’s) health in general? Would you say it was very good, good, fair, bad, or is it very bad?"
In addition, parents reported: mother’s highest educational qualification; housing tenure; family type (traditional, single parent, reconstituted); number of stressful life events; family functioning score; parental anxiety-depression score and age of mother at birth of child.
Statistical Analysis
Analysis used SPSS version 10.1. In order to test for the specific contribution of asthma to behavioral and emotional difficulties, in the absence of any other known health difficulties that might have an impact on adjustment, children with any other organic condition that might affect physical or mental health, for example, heart disease, epilepsy, or diabetes, were excluded from analyses. Independent samples t tests examined differences between children with asthma and those without on the SDQ (parent-, teacher-, and child rated). Chi-square analyses were used to examine the distribution of psychiatric disorder between the two groups. Linear regression was used to assess effect of asthma on SDQ scores, and impact of asthma on psychiatric diagnoses was analyzed using logistic regression. Logistic regression was used to assess combined impact of asthma and child’s general health (rated by the parent), converted into a dichotomous variable, on SDQ subscale and total scores, controlling for demographic and parental mental health variables. No weighting was applied in analyses, because the main questions concerned relationships between variables within the sample.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
For descriptive statistics and comparisons between children with and without asthma, seeTable 1. Independentt tests showed overall that children with asthma were significantly more likely to have elevated parent- and teacher-rated SDQ scores.
|
SDQ Scores and Asthma
The effect of asthma on each SDQ subscale and SDQ total and impact scores was analyzed using linear regression. Asthma was entered as an independent variable along with the following variables: gender of child; general health of child, mother’s highest educational qualification; rented home; stepfamily; three or more life events; single parent family; family functioning score; parental anxiety–depression score; and age of mother at birth of child. These variables were included because prior logistic regression showed them all to be significant independent predictors of psychiatric disorder in this sample to the exclusion of the other measured risk factors such as social class, employment status, and family income levels.
Asthma predicted two of the parent rated scales: conduct problems (unstandardized ß = –0.10, 95% CI = –0.19 to –0.01;t = 2.13, p = .033) and prosocial behavior (unstandardized ß = 0.11, 95% CI = 0.02–0.20; t = 2.42, p = .016) but did not predict any teacher or self-rated SDQ scores. When the general health variable (which potentiality reflected asthma) was removed from the model, however, asthma was found to significantly predict a number of parent and teacher rated scores: parent-rated total symptoms (unstandardized ß = 0.78, 95% CI = 0.49–1.06;t = 5.31, p = .000); parent-rated emotional symptoms (unstandardized ß = 0.33, 95% CI = 0.23–0.43;t = 6.23, p = .000); parent-rated hyperactivity/inattention (unstandardized ß = 0.21, 95% CI = 0.07–0.34;t = 2.99, p = .003); parent-rated peer relationship problems (unstandardized ß = 0.18, 95% CI = 0.04–0.32;t = 3.96, p = .000); parent-rated impact score (unstandardized ß = 0.11, 95% CI = 0.05–0.16;t = 3.57, p = .000); teacher-rated total score (unstandardized ß = 0.51, 95% CI = 0.17–0.85;t = 2.92, p = .003); teacher-rated emotional symptoms (unstandardized ß = 0.18, 95% CI = 0.06–0.29;t = 2.96, p = .003); teacher-rated hyperactivity/inattention (ß = 0.22, 95% CI = 0.06–0.37;t = 2.66, p = .008); teacher-rated peer relationship problems (unstandardized ß = 0.11, 95% CI = 0.01–0.21;t = 1.99, p = .047); and the teacher-rated impact score (unstandardized ß = 0.60, 95% CI = 0.47–0.91;t = 2.81, p = .005). The percentage variance explained by asthma, however, was low, and less than 1% in each of the regression models.
Psychiatric Diagnoses and Asthma
The impact of asthma on psychiatric diagnoses was analyzed using logistic regression. In all of the regressions, asthma was entered as an independent variable with the same variables used in the linear regressions described above. Although the univariate analyses (Table 2) indicated a different distribution of asthma among those children who had a psychiatric diagnosis and those who did not, asthma was the only variable of the 11 entered that did not predict whether the child had a psychiatric diagnosis, nor did it affect anxiety diagnoses, clinical depressive diagnoses, hyperkinetic diagnosis, conduct diagnoses, or the less common psychiatric diagnoses. When the general health variable was removed from the model, however, asthma became a significant predictor of anxiety disorders (OR = 1.38, 95% CI = 1.05–1.82) and hyperkinetic disorders (OR = 1.64, 95% CI = 1.00–2.39). Again, the percentage variance explained by asthma was low, and less than 1% (Nagelkerke pseudo R square) in both logistic regressions.
|
Asthma and General Health
In order to assess the combined impact of asthma and general health on SDQ scores and psychiatric diagnoses, we created a new variable by combining the asthma variable (asthma present or absent) with the general health variable: poor health (very bad/bad/fair) vs. good (good/very good health). This resulted in four new groups:
This new variable was used in a series of logistic regressions with dichotomized SDQ scores (normal vs. borderline or abnormal cases) and psychiatric diagnoses (disorder present/absent) as the outcome variables.Table 3 shows the odds ratios and confidence intervals for asthma/good health, poor health/without asthma, and asthma/poor health in comparison to the healthy children without asthma. Because of the number of comparisons, odds ratios that are significant beyond the 0.01 level are considered significant, but those beyond the 0.001 level can be treated with greater confidence. Findings significant at the 0.05 level are included, but should be interpreted cautiously.
|
Children With Asthma in Good Health
These children showed almost no significantly elevated odds ratios. They were significantly more likely than healthy children without asthma to be rated by parents and teachers as having emotional difficulties, and to be rated by teachers as showing hyperactivity.
Children Without Asthma in Poor Health
Children in poor health without asthma were at significantly greater risk of both parent- and teacher-rated emotional disorder; conduct disorder; hyperactivity; and peer relationship problems. They were also at greater risk of difficulties with teacher-rated prosocial behavior. The children’s own ratings revealed greater risk of both emotional disorder and hyperactivity. DAWBA ratings confirmed that children in poor health were more likely to have an ICD-10 diagnosis of emotional disorder (anxiety or depression) and conduct disorder. They were also more likely to be diagnosed with any psychiatric disorder.
Children With Asthma and in Poor Health
This group had significantly higher odds ratios for all ratings of total difficulties and for parent and teacher ratings of emotional difficulties. The children’s own ratings of emotional difficulties were significant at the 0.05 level and the odds ratio for a diagnosis of emotional disorder also suggested a significant increase in risk. These children were at greater risk of parent-rated conduct disorder, parent-rated hyperactivity, and parent-rated peer relationship problems, and more likely to have ICD-10 diagnoses of conduct disorder and hyperkinetic disorder (hyperactivity). Teachers rated them as having higher peer relationship problems and hyperactivity.
Table 3 gives results of Pearson 
2 analyses comparing without asthma/poor health to with asthma/poor health divided by SDQ category (normal vs. borderline or abnormal). None of the differences was significant at the 0.01 level.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
These are independent effects, with a wide range of other demographic and parental variables controlled for in our regression analyses. It is possible that psychological distress in the informant might contribute to perception of health. For the parent reports, we were able to control for parental depression in our analyses. This was not the case for the child’s self-report, so this reservation should be borne in mind. It should also be noted that in the UK context, provision through the National Health Service ensures free access to health services for all children, so that any potential impact of access to services on the findings should be minimal.
Our findings are consistent with other research indicating that illness severity may be relevant; children with mild to moderate asthma show only modest elevation of psychological difficulties (17). Whether some of the children in our study were in good health because their asthma was less severe, or because it was particularly effectively managed is unknown. Furthermore, because the study is cross sectional, and we do not know the direction of relationships between severity of asthma and psychosocial dysfunction, and whether there are common causal factors, we must be cautious in making recommendations. However, overall, our results are consistent with a model that functional morbidity is associated with psychological risk (10), and indicates directions for targeted psychological intervention.
Our analyses enabled us to explore whether children with asthma were at increased risk of hyperactivity. Although only teachers considered that children with asthma in good health were at increased risk, the odds ratios for children with asthma and in poor health are somewhat higher than they were for children in poor health without asthma. Odds ratios for a diagnosis of hyperactivity were also significantly elevated for children with asthma and in poor health, but not for children with asthma in good health or in poor health without asthma. These differences in odds ratios are noteworthy because, for the majority of ratings, the children with poor health without asthma tended to have equivalent or higher odds ratios to the children with asthma and in poor health. Hyperactivity rated by parents and diagnosed by clinicians is the exception to this.
We have to be cautious in interpreting these findings, because we do not have other external validation of hyperactivity or inattention, and because of the risk that some of our significant findings may be due to chance, given the large number of analyses that we have undertaken. Therefore, our findings cannot convincingly support a pattern of elevated levels of hyperactivity in children with asthma. This is consistent with the findings of a large-scale trial of children with mild and moderate asthma in which neurocognitive functioning was assessed, and which indicated no significant differences in inhibition and attention tasks (9). However, we assume that our population sample includes children with all levels of asthma, and a wide range of levels of medication. There is evidence from other clinical studies that hyperactivity may be related to type of medication prescribed; high-dose oral corticosteroids may affect child behavior (7,8). Because of the nature of our dataset, we could not pursue this line of inquiry, but our findings suggest that there is potential value in studying further whether there are specific factors such as severity of asthma or medication that increase risk of hyperactivity for specific subgroups.
It is beyond the scope of this paper to report the large number of significant findings on the set of other variables also included in the regression analyses. What are reported here are relationships between asthma and psychological adjustment that remain statistically significant when a wide range of other variables are controlled for, although we acknowledge that the percentage variance accounted for by asthma is small. We also acknowledge that the study could not provide detailed information on the child’s physical health status. Ratings were made by one parent, and we had no external validation of this. The level of reporting of asthma, at 15.3%, compares to the Isle of Wight study, where 13.0% of 10-year-olds were currently diagnosed with asthma (18), or the ISAAC study of UK 12- to 14-year-olds (19), where 19.8% self-reported asthma treatment in the past year.
Despite these limitations, our findings represent a replication of previous findings with a very large national epidemiological sample of children. Further research would benefit from closer specification of level of child health when studying psychosocial factors in asthma. Our results are good news for children with asthma who are in good health. They are also good news for family physicians and pediatricians, faced with decision making in resource allocation for this large group of patients, because the results tend to indicate that it would be appropriate to direct any available additional psychosocial resources toward children whose asthma is poorly controlled and who are physically unwell. The levels of distress reported for children who are in poor health and who have asthma indicate that there would be value in further work to elucidate relationships between parenting and illness and to establish services for early identification of, and intervention for psychological difficulties in children at risk of asthma, to minimize the impact of physical illness and psychological disturbance.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
DOI:10.1097/01.psy.0000151490.77622.37
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J.J. Cutuli, J. E. Herbers, M. Rinaldi, A. S. Masten, and C. N. Oberg Asthma and Behavior in Homeless 4- to 7-Year-Olds Pediatrics, January 1, 2010; 125(1): 145 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Glazebrook, A. C. McPherson, I. A. Macdonald, J. A. Swift, C. Ramsay, R. Newbould, and A. Smyth Asthma as a Barrier to Children's Physical Activity: Implications for Body Mass Index and Mental Health Pediatrics, December 1, 2006; 118(6): 2443 - 2449. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
