 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Pediatrics
d Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, Washington
b Children's Hospital and Regional Medical Center, Seattle, Washington
c Center for Health Studies, Group Health Cooperative of Puget Sound, Seattle, Washington
 |
ABSTRACT |
|---|
 TOP ABSTRACT MethodsRESULTSDISCUSSIONREFERENCES |
|---|
METHODS. We conducted a telephone survey of 767 youth with asthma (aged 11–17 years) enrolled in a staff model health maintenance organization. The Diagnostic Interview Schedule for Children was used to diagnose Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, anxiety and depressive disorders; the Child Health Status-Asthma questionnaire (modified) was used to assess asthma symptoms; and automated administrative data were used to measure asthma treatment intensity and severity. Analyses of covariance were performed to determine whether the number of anxiety and depressive symptoms was related to the number of asthma symptoms. Logistic regression analyses were used to evaluate the strength of association between individual symptoms of asthma and the presence of an anxiety or depressive disorder and objective measures of asthma severity.
RESULTS. After adjusting for demographic characteristics, objective measures of asthma severity, medical comorbidity, and asthma treatment intensity, youth with 
1 anxiety or depressive disorder (N = 125) reported significantly more days of asthma symptoms over the previous 2 weeks than youth with no anxiety or depressive disorders. The overall number of reported asthma symptoms was significantly associated with the number of anxiety and depressive symptoms endorsed by youth. In logistic regression analyses, having an anxiety or depressive disorder was also strongly associated with each of the 6 asthma-specific symptoms, as well as the 5 related nonspecific somatic symptoms contained in the Child Health Status-Asthma questionnaire.
CONCLUSIONS. The presence of an anxiety or depressive disorder is highly associated with increased asthma symptom burden for youth with asthma.
Key Words: anxiety • depression • asthma
Abbreviations: ED—emergency department • GHC—Group Health Cooperative • CHSA-T—Child Health Status-Asthma for Teens • DSM-IV—Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition • ANCOVA—analysis of covariance • PCDS—Pediatric Chronic Disease Scale • CI—confidence interval • OR—odds ratio • HEDIS—Health Plan Employer Data and Information Set
Asthma is the most common chronic medical illness of childhood with a prevalence of 3% to 7%.1–3 Among individuals with asthma, there is considerable variation in asthma symptom burden,4–6 which is an important measure used by providers to guide the treatment of asthma and to assess appropriateness of the treatment regimen. The level of symptoms is strongly associated with increased risk of emergency department (ED) visits and hospitalizations,6 as well as decreased quality of life.7,8 These data suggest that it is important to understand what factors are associated with higher symptom burden.
Patient demographic characteristics that have been found to be associated with increased symptom burden include age and gender.6 Modifiable risk factors that have been identified include environmental exposures, exposure to tobacco smoke,6,9 treatment factors (use of inhaled anti-inflammatories10 and use of a written care plan11), and coexisting medical conditions, such as allergic rhinitis and sinusitis.12–18
There is growing evidence that youth with asthma also are at increased risk for anxiety and depressive disorders.19–27 Among adults with asthma, the co-occurrence of an anxiety disorder has been shown to be associated with increased requests for asthma medication,28 increased ED visits,29 and increased hospitalizations.30 However, few studies have examined the association of psychological disorders with asthma symptom burden. A recent study showed that inner-city children with asthma who scored higher on a measure of psychological distress had more hospitalizations, more days of wheezing, and lower functional status than those scoring lower on distress.31 The goal of the current study was to evaluate the association between asthma symptomatology and anxiety/depressive disorder presence and asthma severity in a large population-based sample of adolescents with asthma.
|
Methods |
|---|
|
TOPABSTRACTMethodsRESULTSDISCUSSIONREFERENCES |
|---|
Inclusion Criteria
Potential study subjects were youth (11–17 years of age) with asthma who were enrolled in a GHC insurance plan for 6 months. Administrative data from GHC were used to identify youth with asthma based on meeting 1 of the following criteria: (1) hospitalization in the past year with an asthma diagnosis and 1 asthma prescription during that same time period; (2) 1 ED or urgent care visit for asthma in the past year and 1 asthma prescription during that same time period; (3) 2 office visits for asthma in the past year and 1 asthma prescription during the same time period; (4) 1 office visit for asthma in the past year and another in the past 18 months and 1 asthma prescription in the last year; (5) only 1 asthma visit in the past year but 2 asthma prescriptions filled on different days in that time period; and (6) 4 prescriptions for asthma medication in the last 12 months.
These criteria were developed to identify youth with active asthma and to screen out patients with very mild asthma, such as mild exercise-induced asthma. Youth with an International Classification of Diseases, Ninth Revision code for bipolar disorder or schizophrenia were excluded from the study. All of the youth meeting inclusion criteria were invited to participate in the study.
Study Sample
Of the 1458 children/adolescents and parents in the initial sample, 170 were ineligible, leaving an eligible sample of 1288. Reasons for ineligibility included: child did not have asthma (n = 63), disenrolled from GHC (n = 84), language ineligible (n = 11), parent too ill (n = 6), and other (n = 6). Of the eligible sample, 833 parents gave consent and permission for us to contact their child/adolescent with asthma. From these, we obtained child consent and completed 781 child/adolescent interviews for a final recruitment rate of 60.6%. The final sample for the current analysis is 767 youth (12 youth did not give permission for the use of automated data and 2 youth did not complete the entire psychiatric interview).
Survey Methods
All of the survey contacts were conducted via telephone. At the time of interview, informed consent was obtained both from a parent and the youth study participant. The telephone interview included a 10- to 15-minute interview of the consenting parent and a 45- to 75-minute child/adolescent interview.
Demographics
The parent interview included questions about the child's race/ethnicity, education and employment status for both the responding parent and his/her partner, marital status of the responding parent, and number of children in the household. Child age and gender was obtained from administrative data and confirmed with the parent.
Asthma Diagnosis and Symptom Burden
Asthma diagnosis was identified via administrative data as outlined above and was confirmed with the parent during the parent interview. Medication use was assessed in 2 ways: via the parent interview and administrative data. Treatment intensity was evaluated based on medication fills identified in the administrative data. Youth were classified as receiving no medications, albuterol only, 1 controller, or 2 controllers. Asthma duration was assessed via parent questionnaire and was defined as number of years since diagnosis of asthma.
As a measure of asthma-specific functional status, all of the youth completed a modified version of the Child Health Status-Asthma for Teens (CHSA-T).32 The CHSA-T is a 30-item asthma-specific instrument that has been found to have high reliability and validity in capturing a broad range of asthma experiences.32 The main outcome examined in this article was the number of asthma symptom days reported over the previous 2 weeks. This instrument also includes a 5-point Likert scale, which asks youth to report how much of the time, because of their asthma, they were troubled by each of 11 physical symptoms during the past 2 weeks. Response options ranged from "all of the time" to "none of the time." These 11 symptoms included 6 that are core symptoms of asthma and 5 that are less specific symptoms. The presence or absence of these 11 symptoms was examined as secondary outcomes in our analysis. For the purpose of this analysis, youth were considered to have a symptom if they reported having a symptom "some of the time," "most of the time," or "all of the time" in the previous 2 weeks.
Mental Health Assessment
All of the youth completed the National Institute of Mental Health Diagnostic Interview Schedule for Children (version 4.0), a structured psychiatric interview that has been shown to be a reliable and valid structured interview to diagnose Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), disorders in children and adolescents.33 After standardized training and demonstrated competence, trained staff used the computer-assisted version of the Diagnostic Interview Schedule for Children to guide telephone administration of this instrument. Telephone versions of structured psychiatric interviews have been shown to have a high correlation with in-person interviews.34,35 Interview quality was repeatedly assessed via silent monitors installed on all telephones; interviewers received written feedback on errors, as well as corrective instruction.
For our analysis, youth were considered to have an anxiety (panic, generalized anxiety, separation anxiety, social phobia, or agoraphobia) or depressive (major depression or dysthymia) disorder if they met DSM-IV criteria for 1 of these disorders. Summed scores of the total number (of 62) of positive symptoms across all of the anxiety and depressive disorders were used in the analyses of covariance (ANCOVAs) examining the association between the number of anxiety or depressive symptoms and asthma symptoms.
Asthma Severity
Because symptoms were the key dependent variable in our analyses, we used administrative data to generate a severity measure that was not dependent on reported symptoms. We first explored the use of the Health Plan Employer Data and Information Set (HEDIS) criteria to identify adolescents who were at high risk for adverse asthmatic events. Youth met these criteria if they had any of the following 4 health care use variables over a 12-month period: (1) 4 dispensings of asthma medication; (2) 1 emergency visit for asthma; (3) 1 hospitalization for asthma; or (4) 4 ambulatory visits for asthma.36 In preliminary analysis, we found that 
66% of our sample met these criteria, with most of them meeting criteria because of 4 dispensings of the asthma medication, suggesting lack of specificity as a measure of severity. Therefore, we modified these criteria to develop a more restrictive definition for severity based on administrative data. These modified criteria (having 4 ambulatory visits for asthma, 1 ED visit, 1 hospitalization for asthma, or 1 oral steroid prescription for asthma over a 12-month period) were used to define high-risk youth for all of the analyses in this article.
Medical Comorbidity
The Pediatric Chronic Disease Scale (PCDS) was used to measure health-related medical comorbidity not because of asthma or mental illness.37 The PCDS is an algorithm that classifies children into chronic disease categories by using claims data for prescription fills; studies have shown that it has performed as well as the International Classification of Diseases, Ninth Revision-Clinical Modification-based Ambulatory Care Groups38 in predicting subsequent 1-year health use and health care costs. The PCDS was adapted for our study by removing medications used primarily for asthma, anxiety, and depression.
Statistical Analyses
For the purpose of analysis, youth were categorized as having 1 DSM-IV anxiety or depression diagnosis or no anxiety or depression diagnoses. Basic descriptive statistics were then used to compare the demographic characteristics, asthma severity, asthma duration, treatment factors, and chronic disease score between these 2 groups, using 
2 analyses with corrections for continuity for the categorical variables and t tests for the continuous variables.
A multiple linear regression model was used to examine the relationship between the number of symptom days of asthma in the previous 2 weeks and anxiety or depressive disorders in youth while adjusting for asthma severity, age, gender, parental education, race, ethnicity, duration of asthma, asthma treatment intensity, and PCDS. Because of the small deviation from normality present in the number of symptom days outcome, we conducted a similar analysis using the natural log transformation of number of symptom days (plus 0.5) as the dependent variable with the same set of explanatory variables.
An ANCOVA was used to examine the association between the number of youth-reported symptoms of asthma and the number of youth-reported symptoms of anxiety and depression. Six groups of approximately equal numbers of adolescents were formed based on the number of anxiety and depressive symptoms they reported. This was the independent variable, and the number of asthma symptoms was the dependent variable. Age, gender, parental education, race, ethnicity, asthma duration, asthma treatment intensity, and PCDS were used as covariates.
Because the overall number of symptoms and symptom days was found to be statistically significantly related to anxiety and depressive disorders, logistic regression models were used to examine the association between presence of an anxiety or depressive disorder and asthma severity and each of the asthma-related symptoms in the CHSA-T symptom scale. Variables included in the models were chosen a priori and included age, gender, parental education, race, ethnicity, duration of asthma, asthma treatment intensity, and PCDS. For this analysis, the modified HEDIS asthma severity measure was used as described above. We performed sensitivity analyses for the severity measure by requiring even more stringent criteria of having 2 steroid fills in the previous year or 2 ED visits or 1 hospitalization. However, because the results of the regression analysis were not significantly different with more stringent severity criteria, results are only presented for the main analysis. Based on the a priori hypothesis that youth on controller medications may be different from those not on controller medications, controller medication use was examined as a potential effect modifier. However, the interaction term for this variable was not significant and was not included in the final model.
Because of the fact that we had a final recruitment rate of 60.6%, analyses were all conducted using propensity score weighting. We estimated response propensity scores (probability of being a respondent) as a function of the following variables (all of these within the past year): age, gender, rural-urban commuting area code (defining rural versus urban areas using zip code), being on Medicaid, having state-funded insurance because of low income, PCDS, number of primary care visits, number of asthma-related ED visits and hospitalizations, oral steroid prescription, number of specialty mental health visits, any prescription for antidepressant medication or antianxiety medication, and a diagnosis of depression or anxiety. We predicted response/nonresponse status as a function of these variables using PROC logistic (SAS Institute, Cary, NC) and estimated a response probability for each survey respondent (response propensity score). These propensity scores weights were then applied in all of the analyses (weights inversely proportional to estimated probability of response) after being rescaled to sum to the observed sample size (ie, the number of survey respondents) such that individuals with a low probability of response were given a higher weight in the analysis to represent the larger number of nonrespondents with similar characteristics.
|
RESULTS |
|---|
|
TOPABSTRACTMethodsRESULTSDISCUSSIONREFERENCES |
|---|
1 anxiety and depressive disorders in the last 12 months with 68 (8.9%) having an anxiety disorder alone, 21 (2.5%) a depressive disorder alone, and 37 (4.8%) both an anxiety and depressive disorder. Table 1 describes the demographic characteristics of youth without an anxiety or depressive disorder compared with those with 1 anxiety or depressive disorders. Compared with youth without an anxiety or depressive disorder, youth with a disorder were more likely to be girls, to have a parent with high school education or less, to have a more recent diagnosis of asthma, to be on a single controller and less likely to be on 2 controllers, and to have a higher PCDS score. There were no differences between the 2 groups in either objective measure of severity or other health services use for asthma.
|
2 controller medications. Increasing age was not significantly associated with increased symptom days for these 11- to 17-year-olds. The full model accounts for 6% of the variance, suggesting that other factors not measured account for most of the variance in the outcome. To evaluate the assumption of normal distribution, sensitivity analyses were also conducted using a natural log + 0.5 transformation for the symptom day outcome. In this analysis, the anxiety/depression variable (t = 4.52; P < .001) and PCDS (t = 2.28; P < .03) remained significant predictors; however, gender and treatment intensity were no longer significant.
|
|
|
1 of the following in the previous 12 months: asthma hospitalization, ED visit for asthma, a course of oral steroids, and 4 visits for asthma). Twenty-six percent of youth had 1 of these measures of severity. After controlling for the presence of 1 anxiety or depressive disorder and other potential confounders, youth with more severe asthma based on these objective criteria were significantly more likely to report only 2 of the 11 symptoms from the CHSA-T symptom scale: headache and itchy eyes. As a sensitivity analysis, we examined more restrictive severity criteria for asthma (having 2 ED visits for asthma, 1 inpatient hospitalization, or 2 oral steroid courses for asthma in the previous 12 months). Ten percent of the sample (n = 79) met these more stringent severity criteria. Youth with the more stringent criteria compared with those with less severe asthma were significantly more likely to report tightness in the chest in the previous 2 weeks (OR: 2.02; 95% CI: 1.20 to 3.39) and itchy eyes (OR: 2.00; 95% CI: 1.16 to 3.44) but were not significantly more likely to report other asthma-specific or less-specific physical symptoms.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMethodsRESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge, this is the first study examining the association between anxiety/depressive disorders and asthma symptom burden in a population-based sample of youth with asthma and adds to a growing literature describing a strong relationship between physical symptoms and psychological distress among persons with chronic medical disorders, such as inflammatory bowel disease, diabetes, and coronary artery disease.39 In a recent study of a small school-based sample of adolescents with asthma and other chronic diseases, social phobia was found to be predictive of the number and intensity of respiratory symptoms reported, and other anxiety conditions were found to be associated with the number and intensity of related somatic symptoms.40 These studies suggesting an increase in symptoms for youth with asthma are similar to what has been seen for other chronic diseases. Among adults, the presence of a depressive disorder has been shown to be a stronger predictor of reporting each of 10 diabetes symptoms than the hemoglobin A1C level or the presence of 2 diabetes complications.41 In another study, the presence of depression was more highly associated with chest pain and fatigue than objective measures of coronary artery disease severity.42
There are many potential reasons that youth with anxiety and depressive disorders may report more symptoms of asthma. First, asthma is associated with frequent somatic symptoms, such as breathlessness or chest tightness. Individuals with comorbid mental health disorders have been shown to have more difficulty adapting to aversive symptoms of their diseases.43 Similarly, depressive disorders can be associated with increased focus on illness episodes and medical symptoms.44 This may result in a higher perception and report of symptoms for youth with depressive and anxiety disorders in the absence of objectively measured differences in severity. Given the similarities of symptoms between asthma and anxiety disorders, it is also possible that youth may have difficulty distinguishing between symptoms because of asthma and those because of anxiety. Although youth in our questionnaire were asked specifically about symptoms that they attribute to their asthma, it is not clear whether individuals with asthma and anxiety disorders can accurately distinguish the different causes of their symptoms. Finally, it is possible that, through physiologic mechanisms, the stress related to anxiety and depressive disorders may trigger airway obstruction resulting in more symptoms.45 Because of the cross-sectional nature of these data, we are unable to draw conclusions of causality; however, given the observed associations, assessing for anxiety and depression may be an important part of assessing youth with asthma.
In contrast to the strong association between having an anxiety or depressive disorder and asthma symptom burden, youth meeting objective criteria for more severe asthma were not more likely to report any of the asthma-specific symptoms evaluated and were only more likely to report a few of the nonspecific medical symptoms. Often asthma severity is assessed by frequency of reported symptoms in combination with pulmonary function testing. We did not have pulmonary function testing results on this population. Because symptoms were an outcome for this analysis, we tried to select a measure of severity that was not dependent on symptom reporting. We did explore more stringent measures of severity using administrative data, but they did not substantially change the analysis for anxiety and depression. The lack of association in our study may be because of the fact that our asthma measure is based on health care use, which measures not only disease severity but also individuals seeking treatment for the disease. Treatment seeking may be influenced by many factors other than symptoms, such as ability to cope with symptoms, self-efficacy, and access to care.
It is also possible that youth who had increased health care use for asthma in the previous year were more likely to be treated with a controller medication, which decreased their symptoms at the time of the survey. To assess this possibility, we evaluated whether the association between asthma severity and symptom reporting was different for youth on controller medications versus those not on controller medications. We did not find a significant difference between these 2 groups, suggesting that the lack of association was not because of improved controller medication use. Another possible reason for the lack of association between our asthma severity measure and symptom reporting is the episodic nature of asthma. Our severity measure was based on use in the previous year, whereas the symptom reporting was based on the previous 2 weeks.
Strengths of this study include the large population-based primary care sample of youth with asthma and the use of automated data to identify objective markers of asthma severity and medical comorbidity. It is possible that pulmonary function testing may have had a higher association with symptom reporting than the measure we used. However, given the strong associations between anxiety and depressive disorders and symptom reporting (both asthma-specific symptoms and more general symptoms) and the fact that anxiety and depressive disorders seemed to be only weakly linked to controller medication use and other measures of severity in our analysis, the use of pulmonary function tests as a severity measure would be unlikely to substantially change the observed associations for anxiety and depressive disorders. It is possible that our severity measure was not precise enough to fully measure confounding. To the degree that misclassification occurred, we may have overestimated or underestimated the association between anxiety and depressive disorders and symptom burden. However, the association that we observed is consistent in direction with the one previous study in this area and numerous other studies of the impact of anxiety and depressive disorders on symptom reporting in other chronic diseases, such as diabetes and coronary artery disease.
The other main limitation of the study is that because this study was conducted in a sample from a Northwest group model health maintenance organization, the results may not generalize to populations in other health care settings or different regions of the country.
Health care providers are often challenged by patients who report higher levels of physical symptoms than other patients with comparable disease severity. Physicians report higher levels of frustration with patients who have high numbers of unexplained symptoms.46,47 Depression has also been shown to be associated with differential views of severity of medical concerns between patients and physicians with patients perceiving their symptoms as more severe than clinicians.48 High levels of symptoms that do not correlate with physical findings should prompt providers to assess for anxiety and depression. The recognition of anxiety or depression may decrease physician frustration, decrease the use of unnecessary testing and treatments, and allow physicians to more appropriately target interventions that meet the needs of the patient.
Irrespective of whether anxiety/depressive disorders are the cause or a consequence of asthma symptoms or because of a common underlying factor, recognition and treatment of these mental health disorders among persons with asthma can be expected to improve symptom burden and quality of life. Although this cross-sectional study cannot demonstrate that the treatment of anxiety and depressive disorders will reduce the severity of asthma symptoms, it does suggest that these disorders are associated with a higher likelihood of reporting a broad range of physical symptoms of asthma and medically related symptoms. It will be important to test in future studies whether identifying and treating anxiety and depressive disorders in youth with asthma will result in decreased asthma symptom burden.
|
ACKNOWLEDGMENTS |
|---|
We give special thanks to Michelle Garrison for her editorial assistance.
|
FOOTNOTES |
|---|
Address correspondence to Laura Richardson, MD, MPH, Department of Pediatrics, Box 354920, University of Washington School of Medicine, 6200 NE 74th St, Suite 210, Seattle, WA 98115. E-mail: lpr{at}u.washington.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMethodsRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D C Fuhr and M J De Silva Physical long-term health problems and mental comorbidity: evidence from Vietnam Arch. Dis. Child., August 1, 2008; 93(8): 686 - 689. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
