educational interventions for asthma in children

8/18/2019 Educational Interventions for Asthma in Children

1/187

Educational interventions for asthma in children (Review)

Wolf F, Guevara JP, Grum CM, Clark NM, Cates CJ

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library

2008, Issue 4http://www.thecochranelibrary.com

Educational interventions for asthma in children (Review)

Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

http://www.thecochranelibrary.com/http://www.thecochranelibrary.com/


2/187

T A B L E O F C O N T E N T S

1HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

14CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Analysis 1.1. Comparison 1 Self-management vs. Usual Care, Outcome 1 Lung Function. . . . . . . . . . . 61

Analysis 1.2. Comparison 1 Self-management vs. Usual Care, Outcome 2 Exacerbations (% Patients). . . . . . 62

Analysis 1.3. Comparison 1 Self-management vs. Usual Care, Outcome 3 Exacerbations (Mean). . . . . . . . 62

Analysis 1.4. Comparison 1 Self-management vs. Usual Care, Outcome 4 School Absences (% Patients). . . . . . 63

Analysis 1.5. Comparison 1 Self-management vs. Usual Care, Outcome 5 School Absences (mean days). . . . . . 63

Analysis 1.6. Comparison 1 Self-management vs. Usual Care, Outcome 6 Restricted Activity (% Patients). . . . . 64

Analysis 1.7. Comparison 1 Self-management vs. Usual Care, Outcome 7 Restricted Activity (Mean Days). . . . . 65

Analysis 1.8. Comparison 1 Self-management vs. Usual Care, Outcome 8 Nights Nocturnal Asthma (% Patients). . 65

Analysis 1.9. Comparison 1 Self-management vs. Usual Care, Outcome 9 Nights Nocturnal Asthma. . . . . . . 66

Analysis 1.10. Comparison 1 Self-management vs. Usual Care, Outcome 10 Self-efficacy Scale. . . . . . . . . 66

Analysis 1.11. Comparison 1 Self-management vs. Usual Care, Outcome 11 Asthma Severity Score. . . . . . . 67

Analysis 1.12. Comparison 1 Self-management vs. Usual Care, Outcome 12 General Practitioner Visits. . . . . . 67 Analysis 1.13. Comparison 1 Self-management vs. Usual Care, Outcome 13 ED Visits (% Patients). . . . . . . 68

Analysis 1.14. Comparison 1 Self-management vs. Usual Care, Outcome 14 ED Visits (mean). . . . . . . . . 69

Analysis 1.15. Comparison 1 Self-management vs. Usual Care, Outcome 15 Hospitalization (% patients). . . . . 70

Analysis 1.16. Comparison 1 Self-management vs. Usual Care, Outcome 16 Hospitalizations (mean). . . . . . . 71

Analysis 2.1. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 1 Lung Function. 72

Analysis 2.2. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 2 Exacerbation (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Analysis 2.3. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 3 Exacerbations

(Mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Analysis 2.4. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 4 School Absences (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Analysis 2.5. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 5 School Absences

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Analysis 2.6. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 6 Restricted Activity

(% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Analysis 2.7. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 7 Restricted Activity

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Analysis 2.8. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 8 Nights Nocturnal

Asthma (% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Analysis 2.9. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 9 Nights Nocturnal

Asthma (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Analysis 2.10. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 10 Self-Efficacy

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Analysis 2.11. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 11 Asthma Severity

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

iEducational interventions for asthma in children (Review)



3/187

Analysis 2.12. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 12 General Practitioner

visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Analysis 2.13. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 13 ED Visit (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Analysis 2.14. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 14 ED Visits

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Analysis 2.15. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 15 Hospitalization (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Analysis 2.16. Comparison 2 Self-Management vs. Usual Care by Time Since Enrollment, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

Analysis 3.1. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 1 Lung Function. 85

Analysis 3.2. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 2 Exacerbation (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Analysis 3.3. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 3 Exacerbations

(Mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Analysis 3.4. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 4 School Absences

(% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Analysis 3.5. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 5 School Absences

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Analysis 3.6. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 6 Restricted Activity

(% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Analysis 3.7. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 7 Restricted Activity

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

Analysis 3.8. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 8 Nights Nocturnal

Asthma (% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Analysis 3.9. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 9 Nights Nocturnal

Asthma (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Analysis 3.10. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 10 Self-Efficacy Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Analysis 3.11. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 11 Asthma Severity

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Analysis 3.12. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 12 General

Practitioner visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Analysis 3.13. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 13 ED Visit (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Analysis 3.14. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 14 ED Visits

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Analysis 3.15. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 15 Hospitalization

(% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Analysis 3.16. Comparison 3 Self-management vs. Usual Care by Self-management Strategy, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Analysis 4.1. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 1 Lung Function. . . 99

Analysis 4.2. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 2 Exacerbation (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

Analysis 4.3. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 3 Exacerbations (Mean). 101

Analysis 4.4. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 4 School Absences (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Analysis 4.5. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 5 School Absences (mean

days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Analysis 4.6. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 6 Restricted Activity (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Analysis 4.7. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 7 Restricted Activity (mean

days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

iiEducational interventions for asthma in children (Review)



4/187

Analysis 4.8. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 8 Nights Nocturnal Asthma

(% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Analysis 4.9. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 9 Nights Nocturnal Asthma

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Analysis 4.10. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 10 Self-Efficacy Scale. 106

Analysis 4.11. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 11 Asthma Severity

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Analysis 4.12. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 12 General Practitioner

visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Analysis 4.13. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 13 ED Visit (% patients). 109

Analysis 4.14. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 14 ED Visits (mean). 110

Analysis 4.15. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 15 Hospitalization (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Analysis 4.16. Comparison 4 Self-management vs. Usual Care by Intervention Type, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Analysis 5.1. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 1 Lung Function. . 113

Analysis 5.2. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 2 Exacerbation (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Analysis 5.3. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 3 Exacerbations

(Mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Analysis 5.4. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 4 School Absences (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Analysis 5.5. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 5 School Absences (mean

days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Analysis 5.6. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 6 Restricted Activity (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Analysis 5.7. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 7 Restricted Activity

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Analysis 5.8. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 8 Nights Nocturnal

Asthma (% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Analysis 5.9. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 9 Nights Nocturnal

Asthma (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Analysis 5.10. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 10 Self-Efficacy

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Analysis 5.11. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 11 Asthma Severity

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Analysis 5.12. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 12 General Practitioner

visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

Analysis 5.13. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 13 ED Visit (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Analysis 5.14. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 14 ED Visits (mean). 122 Analysis 5.15. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 15 Hospitalization (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

Analysis 5.16. Comparison 5 Self-management vs. Usual Care by Intervention Intensity, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

Analysis 6.1. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 1 Lung Function. . . . . . 125

Analysis 6.2. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 2 Exacerbation (% patients). . 126

Analysis 6.3. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 3 Exacerbations (Mean). . . 127

Analysis 6.4. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 4 School Absences (% patients). 127

Analysis 6.5. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 5 School Absences (mean days). 128

Analysis 6.6. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 6 Restricted Activity (% patients). 129

Analysis 6.7. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 7 Restricted Activity (mean days). 130

iiiEducational interventions for asthma in children (Review)



5/187

Analysis 6.8. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 8 Nights Nocturnal Asthma (%

Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Analysis 6.9. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 9 Nights Nocturnal Asthma

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Analysis 6.10. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 10 Self-Efficacy Scale. . . . 132

Analysis 6.11. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 11 Asthma Severity Scale. . 133

Analysis 6.12. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 12 General Practitioner visits

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Analysis 6.13. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 13 ED Visit (% patients). . 135

Analysis 6.14. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 14 ED Visits (mean). . . . 136

Analysis 6.15. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 15 Hospitalization (% patients). 137

Analysis 6.16. Comparison 6 Self-management vs. Usual Care by Trial Type, Outcome 16 Hospitalizations (mean). . 138

Analysis 7.1. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 1 Lung

Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Analysis 7.2. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 2Exacerbation (% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

Analysis 7.3. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 3

Exacerbations (Mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Analysis 7.4. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 4 School

Absences (% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Analysis 7.5. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 5 School

Absences (mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142


Restricted Activity (% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . 143


Restricted Activity (mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

Analysis 7.8. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 8 Nights

Nocturnal Asthma (% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Analysis 7.9. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 9 Nights

Nocturnal Asthma (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Analysis 7.10. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 10 Self-

Efficacy Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146


Asthma Severity Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147


General Practitioner visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Analysis 7.13. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 13 ED

Visit (% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Analysis 7.14. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 14 ED

Visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

Analysis 7.15. Comparison 7 Self-management vs. Usual Care by Adequacy of Allocation Concealment, Outcome 15Hospitalization (% patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151


Hospitalizations (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

Analysis 8.1. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 1 Lung Function. 153

Analysis 8.2. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 2 Exacerbation (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Analysis 8.3. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 3 Exacerbations

(Mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Analysis 8.4. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 4 School Absences (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Analysis 8.5. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 5 School Absences

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

ivEducational interventions for asthma in children (Review)



6/187

Analysis 8.6. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 6 Restricted Activity (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Analysis 8.7. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 7 Restricted Activity

(mean days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Analysis 8.8. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 8 Nights Nocturnal

Asthma (% Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

Analysis 8.9. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 9 Nights Nocturnal

Asthma (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

Analysis 8.10. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 10 Self-Efficacy

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

Analysis 8.11. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 11 Asthma Severity

Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

Analysis 8.12. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 12 General Practitioner

visits (mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Analysis 8.13. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 13 ED Visit (%patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Analysis 8.14. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 14 ED Visits

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Analysis 8.15. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 15 Hospitalization (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

Analysis 8.16. Comparison 8 Self-management vs. Usual Care by Adequacy of Follow-up, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

Analysis 9.1. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 1 Lung Function. . . . 166

Analysis 9.2. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 2 Exacerbation (% patients). 167

Analysis 9.3. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 3 Exacerbations (Mean). 168

Analysis 9.4. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 4 School Absences (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Analysis 9.5. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 5 School Absences (meandays). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Analysis 9.6. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 6 Restricted Activity (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

Analysis 9.7. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 7 Restricted Activity (mean

days). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Analysis 9.8. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 8 Nights Nocturnal Asthma (%

Patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

Analysis 9.9. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 9 Nights Nocturnal Asthma

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

Analysis 9.10. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 10 Self-Efficacy Scale. . 173

Analysis 9.11. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 11 Asthma Severity Scale. 174

Analysis 9.12. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 12 General Practitioner visits

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Analysis 9.13. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 13 ED Visit (% patients). 176

Analysis 9.14. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 14 ED Visits (mean). . 177

Analysis 9.15. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 15 Hospitalization (%

patients). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Analysis 9.16. Comparison 9 Self-management vs. Usual Care by Asthma Severity, Outcome 16 Hospitalizations

(mean). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

179 WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

179HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vEducational interventions for asthma in children (Review)



7/187

[Intervention Review]

Educational interventions for asthma in children

Fredric Wolf 1, James P Guevara 2, Cyril M Grum3, Noreen M Clark 4, Christopher J Cates5

1Department of Medical Education & Biomedical Informatics, University of Washington School of Medicine, Seattle, WA, USA.2Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA, USA. 3University of Michigan, Ann

Arbor, MI 48109-0368, USA. 4 School of Public Health, University of Michigan, Ann Arbor, MI 48109-2029, USA. 5 Community

Health Sciences, St George’s, University of London, London, UK

Contact address: Fredric Wolf, Department of Medical Education & Biomedical Informatics, University of Washington School of

Medicine, E-312 Health Sciences, Box 357240, Seattle, WA, 98195-7240, USA. [email protected].

Editorial group: Cochrane Airways Group.Publication status and date: Edited (no change to conclusions), published in Issue 4, 2008.

Review content assessed as up-to-date: 31 July 2002.

Citation: Wolf F, Guevara JP, Grum CM, Clark NM, Cates CJ. Educational interventions for asthma in children. Cochrane Database of Systematic Reviews 2002, Issue 4. Art. No.: CD000326. DOI: 10.1002/14651858.CD000326.


A B S T R A C T

Background

Self-management education programs have been developed for children with asthma, but it is unclear whether such programs improveoutcomes.

Objectives

To determine the efficacy of asthma self-management education on health outcomes in children.

Search strategy

Systematic search of the Cochrane Airways Group’s Special Register of Controlled Trials and PSYCHLIT, and hand searches of the

reference lists of relevant review articles.

Selection criteria

Randomized and controlled clinical trials of asthma self-management education programs in children and adolescents aged 2 to 18

years.

Data collection and analysis

All studies were assessed independently by two reviewers. Disagreements were settled by consensus. Study authors were contacted for

missing data or to verify methods. Subgroup analyses examined the impact of type and intensity of educational intervention, self-

management strategy, trial type, asthma severity, adequacy of follow-up, and study quality.

Main results

Of 45 trials identified, 32 studies involving 3706 patients were eligible. Asthma education programs were associated with moderate

improvement in measures of airflow (standardized mean difference [SMD] 0.50, 95% confidence interval [CI] 0.25 to 0.75) and self-

efficacy scales (SMD 0.36, 95% CI 0.15 to 0.57). Education programs were associated with modest reductions in days of school absence

(SMD -0.14, 95% CI -0.23 to -0.04), days of restricted activity (SMD -0.29, 95% CI -0.49 to -0.08), and emergency room visits

(SMD -0.21, 95% CI -0.33 to -0.09). There was a reduction in nights disturbed by asthma when pooled using a fixed-effects but

not a random-effects model. Effects of education were greater for most outcomes in moderate-severe, compared with mild-moderate

1Educational interventions for asthma in children (Review)


mailto:[email protected]:[email protected]


8/187

asthma, and among studies employing peak flow versus symptom-based strategies. Effects were evident within the first six months, but

for measures of morbidity and health care utilization, were more evident by 12 months.

Authors’ conclusions

Asthma self-management education programs in children improve a wide range of measures of outcome. Self-management education

directed to prevention and management of attacks should be incorporated into routine asthma care. Conclusions about the relative

effectiveness of the various components are limited by the lack of direct comparisons. Future trials of asthma education programs should

focus on morbidity and functional status outcomes, including quality of life, and involve direct comparisons of the various components

of interventions.

P L A I N L A N G U A G E S U M M A R Y

Educational interventions for asthma in children

Learning self-management strategies related to asthma prevention or attack management can help improve children’s lung function

and feelings of self-control, as well as reduce school absences and days of restricted activity and decrease emergency room utilization.

There were no differences in the risk or frequency of hospitalizations between usual care and care supplemented with self-management

education. These types of more rare andseriousevents maybe beyondthe ability of educationto influence. While more research is needed

to make direct comparisons between different types of interventions, the limited evidence currently available suggests that in general,

self-management education works well for persons with moderate-to-severe asthma as well as for those with mild-to-moderate asthma.

Peak flow-based educational strategies generally show greater effects than symptom-based strategies. Beneficial effects on measures of

physiological function were apparent within six months, but benefits did not become fully apparent on measures of morbidity or health

care utilization until 7 to 12 months following enrolment in an educational program.

B A C K G R O U N D

Asthma is the most prevalent chronic pulmonary disorder afflict-

ing children (Shamssain 1999; Habbick 1999; McFadden 1992).

The prevalence of diagnosed asthma has been growing over the

past 20 years (Magnus 1997; Senthilselvan 1998). Of greater con-

cern is that both asthma morbidity and mortality appear to be

increasing (Ng 1999). This increase is seen particularly in lower

socioeconomic groups and in minority populations (Cunningham

1996). Children with asthma suffer a high number of school ab-

sences (Doull 1996), endure a high and increasing rate of disabil-ity (Perrin 1999; Newacheck 2000), and incur substantial health

care costs (Lozano 1997).

Self-management educational programs for children with asthma

have been developed in recent years in recognition of the need to

improve health care practices, reduce morbidity, and lower costs of

care(Hurd 1992; Clark 1989). To be successful, programs must be

basedon a soundtheoretical understandingof behavior change and

employ self-management strategies designed to improve knowl-

edge, skills, and feelings of self-control (Clark 1994). A number

of educational programs have been the subject of rigorous eval-

uations (Lewis 1984; Wilson-Pessano 1985; Clark 1986a ; Clark

1986b; Evans 1987; Creer 1976; Hindi-Alexander 1984; Parcel

1980). These programs incorporate a variety of educational strate-

gies, are designed for different clinical settings, and are targeted to

different patient groups.

While it has become increasingly clear that limited asthma edu-

cation involving only information transfer is ineffective (Gibson

1999), the effectiveness of self-management education programs

in children with asthma is unclear. In adults with asthma, the use

of a self-management education program that includes self-mon-

itoring, regular medical review, and an asthma action plan doesappear to improve measures of morbidity and reduce health care

utilization (Gibson 1998). Inchildren,however, a publishedmeta-

analysis of 11 self-management teaching programs concluded that

asthma self-management programs do not reduce morbidity or

decrease health care utilization (Bernard-Bonnin 1995). This re-

searchsynthesiswas limited to trialspublished prior to 1992, and a

large number of studies have been published subsequently. There-

fore, a new review that incorporates more recent studies may help

to clarify the uncertainty regarding the effectiveness of asthma ed-

ucation in children.

The purpose of this study was to systematically review the research

literature on the efficacy of self-management educational inter-




9/187

ventions in modifying health outcomes for children with asthma.

We hypothesized that self-management programs would be asso-ciated with improvements in measures of lung function, decreases

in measures of morbidity, and lead to reductions in health care

utilization.

O B J E C T I V E S

The specific study objectives were two-fold:

(1) To determine the effectiveness of self-management education

programs on measures of physiological function, morbidity and

functional status, self-perception, and health care utilization in

children and adolescents with asthma.

(2) To determine the characteristics of self-management educa-

tion programs and trials that are associated with improvements in

health outcomes in children and adolescents with asthma.

M E T H O D S

Criteria for considering studies for this review

Types of studiesRandomized controlled trials (RCTs) and controlled clinical trials

(CCTs) were considered in this systematic review.

Types of participants

Studies of children and adolescents with asthma from two to 18

years of age were included.

Types of interventions

Any educational intervention targeted to children or adolescents

(and/or their parents) designed to teach one or more self-manage-

ment strategies related to prevention, attack management, or socialskills using any instructional strategy or combination of strategies

(problem solving, role-playing, videotapes, computer assisted in-

struction, booklets, etc.) presented either individually or in group

sessions was included in the review.

Types of outcome measures

Categories of outcomes examined for this review are based on a

consensus of clinically relevant outcomes from the Asthma Out-

comes Conference and adapted from Clark and Starr-Schneid-

kraut’s model of patient management (Clark 1994). These out-

come categories are (1) physiological function, (2) morbidity and

functional status, (3) self-perception measures, and (4) health care

utilization. Two additional outcome categories from the confer-ence, adverse effects of medications and quality of life, are not in-

cluded here because they were not reported in any eligible trial. A

full listing of more specific outcomes within each category is pro-

vided below. We have selected one outcome from each category as

primary outcomes. These decisions were typically made because

the most data were available for these outcomes, both in terms of

the number of studies and sample sizes for the individual studies

and because these outcomes were considered the most salient in

each category. The primary outcomes are a combined measure of

lung function (1c), days of school absence (2b), self-efficacy (3b),

and emergency department visits (4b).

(1) Physiological function

(a) forced expiratory volume in 1 second (FEV1): exact numberor percent of predicted

(b) peak expiratory flow rate (PEF): exact number or percent of

predicted

(c) a combined measure of lung function: defined as either FEV1

or PEF

(2) Morbidity and functional status

(a) exacerbations: defined as asthma attacks or episodes of asthma

(b) days of school absence: defined as days of school non-atten-

dance due to asthma or other causes

(c) days of restricted activity: defined as days of asthma symptoms

or days of activity restriction

(d) nights disturbed by asthma: defined as nights of sleep inter-

ruption due to asthma or nights with asthma symptoms(3) Self-perception measures

(a)asthma severity: defined as subjective measures of asthmasever-

ity and includes asthma symptom or severity scores

(b) self-efficacy: “Self-efficacy is the belief in one’s capabilities to

organize and execute the sources of action required to manage

prospective situations” (Bandura 1996). A strong sense of personal

efficacy has been shown to be related to a variety of outcomes,

including better health, higher achievement, and more social inte-

gration (Schwarzer 1995). For purposes of this systematic review,

we have included measures of coping scores and health locus of

control scales.

(4) Health care utilization

(a) general practitioner visits: defined as routine and urgent am-bulatory clinic visits to a general practitioner, family physician,

pediatrician, or other related health care provider

(b) emergency department visits: defined as urgent visits to a hos-

pital emergency department

(c) hospitalizations: defined as any inpatient hospital stay

If outcomes were reported separately by time, the outcome cor-

responding to six months or later post-enrollment was selected.

Regarding measures of lung function, if outcomes were reported

at multiple times during the day, the morning measurement was

selected.




10/187

Search methods for identification of studies

Electronic searches

We identified studies from the Cochrane Airways Group’s tri-

als register comprised of references from MEDLINE, EMBASE,

CINAHL, and hand searched airways-related journals. This

database was searched using the following terms:

asthma OR wheez* AND education* OR self management OR

self-management.

The Airways Group databases combine EMBASE, CINAHL and

MEDLINE records. We identified the EMBASE records through

keyword or text word (in title or abstract fields) searches on

asthma* or on the term wheez*. In Medline and CINAHL, the

text word searches (in title or abstract fields) were made on

terms asthma* or wheez*. The MeSH searches in MEDLINE and

CINAHL were made on the following two MeSH terms: respira-

tory sounds and asthma. The MEDLINE records cover the period

1966 to 1998, EMBASE records span 1980 to 1998, and the ear-

liest CINAHL records date from 1982. The main Airways Group

databases include the records downloaded from the three elec-

tronic systems. We searched the main Airways Group databases

using the following terms: placebo* OR trial* OR random* OR

double-blind OR double blind OR single-blind OR single blind

OR controlled study OR comparative study) and we exported

those records to a separate RCT register.

In addition, we identified asthma-related studies from PSYCH-

LIT using the Cochrane Schizophrenia Group’s search strategy to

identify trials in that behavioral sciences database. We searched

this database using the following terms: “asthma* OR asthma- in

DE OR wheez* OR [(bronchial*) near (hyper-reactiv* or hyper-

reactiv*)] AND randomi* OR [(singl* OR doubl* OR trebl* OR

tripl*) near (blind* OR mask*)] OR crossover”.

Searching other resources

Finally, we hand searched the reference lists from relevant re-

views that had been identified (Clark 1993; Clark 1994; Bernard-

Bonnin 1995).

Data collection and analysis

Selection of studies

One reviewer screened the title and abstract of each citation that

we identified through the search strategy to determine possible

eligibility for inclusion. We then obtained the complete article

of each citation identified as eligible or possibly eligible. At least

two investigators (FW, CG, JG, MR) independently assessed each

article to determine study eligibility. Disagreement was settled by

consensus.

Studies were included if (a) they were published randomized con-

trolled trials (RCTs) or controlled clinical trials (CCTs); (b) they includedchildren or adolescents ages 2 to 18 years old; (c)they had

an educational intervention designed to teach one or more self-

management strategies related to prevention, attack management,

or social skills; (d) they included outcomes on pulmonary func-

tion tests, morbidity, functional status, or health care utilization.

We excluded studies if they included participants with pulmonary

diagnoses other than asthma, lacked suitable control populations

for comparison, used non-standard educational interventions, or

did not report on any outcomes of interest.

Data extraction and management

We abstracted all eligible studies onto preprinted data collection

forms. Information on randomization methods, participants, fol-

low-up procedures, nature of educational interventions, and out-

comes was collected. We also contacted, by mail, authors of all

eligible studies to verifythe accuracy of published data or to obtain

missing data. Twenty authors (63%) provided additional informa-

tion on allocation concealment procedures or missing information

that was then incorporated into this review.

Assessment of risk of bias in included studies

We based the methodological quality of included trials primarily

on an assessment of allocation concealment. Allocation conceal-ment refers to whether trials sufficiently concealed group alloca-

tion prior to randomization andwas measured using theCochrane

approach (Clarke 1999):

• Category A: Adequate concealment using formal

randomization procedures (e.g. sealed envelopes or random

numbers).

• Category B: Uncertain or unclear concealment.

• Category C: Clearly inadequate concealment using non-

random procedures (e.g. alternation).

Methodological quality of included trials was also based on assess-

ments of studies indicating whether performance bias, exclusion

bias, or detection bias were present. Performance bias refers to

whether differences in care may have existed between treatment

and control groups apart from the intervention and was measured

as present or absent. Exclusion bias refers to whether significant

differences in withdrawal existed between treatment and control

groups (generally at least 10%) or whether withdrawal rates in the

combined group were large (generally greater than 20%) and was

measured as present or absent. Detection bias refers to whether

differences in outcome assessment may have existed between treat-

ment and control groups and was measured as present or absent.

Dealing with missing data




11/187

In addition, if continuous outcomes were reported without stan-

dard deviations, we imputed pooled standard deviations from thefollowing generalized formula for the t-statistic (Rosenthal 1991):

t = [M1 - M2 / S] X [ 1/ sq root (1/n1 + 1/n2 ) ]

Where t refers to the t-statistic, M refers to the mean of the treat-

ment and control groups, S refers to the pooled standard devia-

tion, and n refers to the sample size of the treatment and control

groups. If the t-statistic was not reported in a given paper, the t-

statistic corresponding to the exact p-value with the appropriate

degrees of freedom was used. If the t-statistic and exact p-value

were not reported in a given paper, the t-statistic corresponding

to p = 0.05 (for a reported p 0.05) with the appropriate degrees of freedom was used.

Data synthesis

For continuous outcomes, we used the standardized weighted

mean difference (SMD) to estimate a pooled effectsize, since in all

cases, outcomes were reported in different units or scales. We re-

ported the pooled effect sizes with 95% confidence intervals (CI).

We used both a fixed-effects model (Hasselblad 1995) and a ran-

dom-effects model (DerSimonian 1986) to pool the data, since

there does not appear to be general consensus as to which method

is superior. In general, the random-effects method is a more con-

servative approach than the fixed-effects method and results in

larger CIs.

For consistency and simplicity, We reported the SMDs based on

the fixed-effects model in the text of the review and in the table of

comparisons. Where effect sizes differed statistically, we reported

the pooled effect size from the random-effects model in footnotes

to the tables and in the text of the review. For dichotomous out-

comes, we used the odds ratio (OR) with 95% CIs to estimate

a pooled effect size (Greenland 1985). We also used both fixed-

effects and random-effects models to pool data on dichotomous

outcomes. We calculated the number needed to treat (NNT) to

prevent an adverse event for all statistically significant dichoto-

mous outcomes ( Altman 1998).

Where necessary, we computed standard deviations from standard

errorsof the mean or confidence intervals using standard statistical

formulas (Hedges 1985; Wolf 1986).

Subgroup analysis and investigation of heterogeneity

We performed subgroup analyses by stratifying studies on key pa-

tient-level, study quality, and program variables in order to esti-

mate the magnitude of these effects.

(1) Patient characteristics: Subgroup analysis was performed on

the effect of educational interventions by asthma severity. Asthma

severity (mild-moderate vs. moderate-severe) was determined by

study self-report, examination of mean FEV1 or PEFR baseline

measurements, or chronicity of asthmasymptomsat baseline ( ATS

1991; NAEP 1997). Patients were determined to have severe

asthma if they had mean FEV1 < 0.50 of predicted, mean PEFR

< 0.60 of predicted, or reported daily asthma symptoms. We cat-egorized studies as moderate to severe if asthmatics with severe

asthma were enrolled in the study population and mild-moderate

otherwise.

(2) Educational program characteristics: Subgroup analyses were

performed on the effect of educational interventions by (a) inter-

vention type (individual vs. group); (b) intensity of intervention

(single vs. multiple sessions); (c) time since enrolment (one to six

months vs. seven to twelve months vs. > twelve months); and (d)

self-management strategy (peakflow-based vs. symptom-based) to

estimate the effect of various intervention characteristics.

(3) Study quality characteristics: Subgroup analyses were per-

formed on the effect of educational interventions by (1) trial type(RCT vs. CCT); (2) allocation concealment (adequate vs. unclear

vs. inadequate); and (3) adequacy of follow-up (adequate vs. in-

adequate) to estimate the effects of study characteristics.

R E S U L T S

Description of studies

See: Characteristicsof included studies; Characteristicsof excluded

studies.

Results of the search

The search identified 318 titles and abstracts of potentially eligi-

ble studies. After preliminary review, a total of 45 randomized or

controlled clinical trials were identified from the literature search

and review of bibliographies as possibly eligible for inclusion. Af-

ter review of the full text of these studies, 13 were excluded for

the following reasons: absence of a suitable control population (N

= 2), inclusion of children less than two years old (N = 2), use of

non-standard or information only educational interventions (N =

4), inclusion of children with conditions other than asthma (N =

1), and no outcomes of interest (N = 4).

Included studies

A total of 32 trials involving 3706 children and adolescents with

asthma were selected for inclusion (see Characteristics of included

studies). Twenty-six (81%) of these trials were RCTs, and the re-

mainder were non-randomized CCTs. The trials varied in size

(mean 116, range 20 to 451 participants), severity of asthma

among participants (15 with moderate-to-severe asthma, four with

mild-to-moderate asthma, and 13 with unclear severity), and the

proportion of participants with complete follow-up (range 43

to 100%). Fifteen of the trials reported inclusion of adolescents




12/187

(age 13 years old and above), while twelve reported inclusion of

preschool age children (ages two to five years old). The self-man-agement educational programs employed by the trials differed by

type of educational session (15 employing group sessions, 14 em-

ploying individual sessions, three employing both), intensity (five

employing only a single session, two employing two sessions, 25

employing three or more sessions), self-management strategy (13

employing peak flow-based strategies, 19 employing symptom-

based strategies), and length of the intervention (mean 3.8, range

one to twelve months). While thirteen trials incorporated social

skills development into their educational strategy, all trials focused

on asthma prevention measures (e.g. identification and avoidance

of asthma triggers) and/or attack management plans (e.g. use of

an asthma action plan).

Outcomes assessed

(1) Measures of physiological function:

Forced expiratory volume in 1 second (FEV1) (N = 2)

Peak expiratory flow rate (PEF) (N = 4)

Combined measure of lung function (N = 7)

(2) Measures of morbidity and functional status:

Proportion with exacerbations (N = 2)

Exacerbations (N = 5)

Proportion with School Absences (N = 1)

Days of School absence (N = 17)

Proportion with restricted activity (N = 1)

Days of restricted activity (N = 6)Proportion with Nights disturbed by asthma (N = 1)

Nights disturbed by asthma (N = 3)

(3) Measures of self-perception:

Self-efficacy scales (N = 10)

Asthma severity scores (N = 5)

(4) Measures of health care utilization:

Physician visits (N = 10)

Proportion with ED visits (N = 6)

ED visits (N = 14)

Proportion with hospitalization (N = 4)

Hospitalizations (N = 9)

Risk of bias in included studies

The methodological quality of the studies varied. Twelve trials

were felt to have adequate concealment of allocation of patients

to intervention or control groups, while six trials had clearly in-

adequate or no concealment methods (eg allocation based on al-

ternating sites or sequential assignment). In fourteen trials, it was

not possible to determine the method of allocation concealment

based on published methods. Six studies had differences in the

care provided to treatment and control groups apart from the in-

tervention. Eight studies had systematic differences in withdrawal

between treatment and control groups. There did not appear to

be systematic differences in the assessment of outcomes between

treatment and control groups in any of the studies.

Effects of interventions

AUTHOR VERIFICATION

An attempt was made to contact corresponding authors of all

studies in order to verify allocation concealment procedures and

to obtain missing data. A total of twentyauthors (63%) responded

to requests for additional information.

OUTCOMES: SELF-MANAGEMENT VERSUS

USUAL CARE (N = 32 trials)

(1) PHYSIOLOGICAL FUNCTION

Four trials involving 258 patients reported complete data on the

effect of self-management education programs on measures of

physiological function. There was a significant improvement on

a combined measure of lung function (SMD 0.50, 95% CI 0.25

to 0.75) as well as on individual measures of FEV1 (SMD 0.46,

95% CI 0.08 to 0.84) and PEF (SMD 0.53, 95% CI 0.19 to

0.86) associated with self-management education programs when

the trials were pooled. Three additional trials involving 192 pa-tients, which could not be pooled due to missing data, reported

no significant effect of education on measures of lung function

(Hughes 1991; Dahl 1990; Szczepanski 1996). The pooled esti-

mates obtained by the random-effects model were consistent with

those from the fixed-effects analyses. There was no significant het-

erogeneity among the trials reporting on the combined measure of

lung function (Chi-square = 2.44; p = 0.49) and PEF (Chi-square

= 2.38; p = 0.3).

(2) MORBIDITY AND FUNCTIONAL STATUS

Eighteen trials involving 1649 patients reported complete data on

the effect of self-management education programs on measures of morbidity and functional status. There was a small but significant

reduction in the days of school absence (SMD -0.14, 95% CI -

0.23 to -0.04) and days of restricted activity (SMD -0.29, 95% CI

-0.49 to -0.08) when the 16 and six trials respectively reporting on

these outcomes were pooled by either the fixed effect or random-

effects model. One additional trial involving 84 patients which

could not be pooled, due to missing data, reported statistically sig-

nificantly less school absenteeism (p < 0.05) among the education

group than the control group (Szczepanski 1996). There was a

significant reduction in the nights disturbed by asthma when the

three trials reporting on this outcome were pooled by the fixed-

effect model (SMD -0.34, 95% CI -0.62 to -0.05) but not by




13/187

the random-effects model (SMD -0.39, 95% CI -1.07 to 0.28).

There was no significant reduction in the number of exacerbations(SMD -0.21, 95% CI -0.43 to 0.01) when the five trials reporting

on this outcome were pooled. In addition, there was no significant

reduction in the proportion of patients experiencing an exacer-

bation (OR 1.43, 95% CI 0.94 to 2.18), day of school absence

(OR 0.78, 95% CI 0.36 to 1.66), day of restricted activity (OR

2.51, 95% CI 0.61 to 10.29), or night disturbed by asthma (OR

0.65, 95% CI 0.29 to 1.44). However, few studies reported on

dichotomous morbidity outcomes. There was significant hetero-

geneity among the trials pooled for nights disturbed by asthma

(Chi-square = 11.19, p = 0.004), but not among trials pooled for

exacerbations, days of school absence, or days of restricted activity.

(3) SELF PERCEPTION

Nine trials involving 522 patients reported complete data on the

effect of self-management educationprogramson measures of self-

perception. There was improvement on measures of self-efficacy

(SMD 0.36, 95% CI 0.15 to 0.57) when the six trials reporting

this outcome were pooled using the fixed effect or random effects

model. Four additional trials that could not be pooled, due to

missing data, reported that education was beneficial on scales of

self-efficacy in three studies involving 313 patients (Evans 1987;

Kubly 1984; Szczepanski 1996) but not in one study involving

43 patients (Rakos 1985). There was no significant reduction by

education on asthma severity scores (SMD -0.15, 95% CI -0.43

to 0.12) when the four trials were pooled. One additional trial

involving 84 patients which could not be pooled due to missing

data reported a decrease in asthma severity scores in the education

plus follow-up group (p < 0.05) but not in the education only

or control groups (Szczepanski 1996). There was no significant

heterogeneity among the trials pooled for self-efficacy scales but

there was heterogeneity among trials pooled for asthma severity

scores (Chi-square = 6.72, p = 0.08).

(4) HEALTH CARE UTILIZATION

Eighteen trials involving 1899 patients reported complete data on

the effect of self-management education programs on measures of health care utilization. There was a significant reduction in the

number of emergency department (ED) visits (SMD -0.21, 95%

CI -0.33 to -0.09) associated with education when the 12 tri-

als reporting complete data on this outcome were pooled using

both fixed effect and random effects models. Two additional tri-

als which could not be pooled due to missing data reported no

benefit of education in one study involving 43 patients (Rakos

1985) but a benefit of education in one study involving 84 pa-

tients (Szczepanski 1996). There was no significant reduction in

the number of general practitioner visits (SMD -0.15, 95% CI -

0.31 to 0.01) when the six trials reporting complete data on this

outcome were pooled. Four additional trials that could not be

pooled due to missing data reported a significant benefit of ed-

ucation on general practitioner visits in one study involving 78patients (Colland 1993) but not in three studies involving 141

patients (McNabb 1985; Rakos 1985; Szczepanski 1996). There

was no significant reduction in the number of hospitalizations

associated with education (SMD -0.08, 95% CI -0.21 to 0.05)

when the eight trials reporting complete data on this outcome were

pooled. One additional trial involving 84 patients which could not

be pooled due to missing data reported no significant reduction

in hospitalizations associated with education (Szczepanski 1996).

Moreover, there was no significant reduction in the proportion of

patients who experienced an ED visit (OR 1.30, 95% CI 0.93

to 1.84) or a hospitalization (OR 1.00, 95% CI 0.70 to 1.42)

when the six trials and four trials respectively reporting data on

these outcomes were pooled. There was significant heterogeneity among the trials pooled for hospitalizations (Chi-square = 13.03,

p = 0.07) and ED visits (Chi-square = 19.68, p = 0.05) but not

for general practitioner visits.

SUBGROUP ANALYSES

OUTCOMES: SELF-MANAGEMENT VERSUS USUAL

CARE BY TIME SINCE ENROLLMENT (N = 32 TRIALS)

When self-management education programs were examined by

time since enrollment, effects of education versus usual care on

measures of physiological function were evident within the firstsix months of enrollment ((SMD 0.50, 95% CI 0.25 to 0.75).

However, effects on measures of morbidity [exacerbations (SMD

-0.28, 95% CI -0.53 to -0.03), days of school absence (SMD -

0.16, 95% CI -0.29 to -0.04), days of restricted activity (SMD

-0.26, 95% CI -0.48 to -0.04), and nights disturbed by asthma

(SMD -0.86, 95% CI -1.38 to -0.35)], self-perception [self-effi-

cacy scales (SMD 0.54, 95% CI 0.28 to 0.80)], and health care

utilization [ED visits (SMD -0.19, 95% CI -0.32 to -0.05), and

hospitalizations (SMD -0.26, 95% CI -0.44 to -0.08)] became

more evident at seven to twelvemonths post-enrollment.Evidence

for effectiveness from longer term follow-up was sparse.


CARE BY SELF-MANAGEMENT STRATEGY (N = 32

TRIALS)

When comparing the effectiveness of self-management strategies,

studies that employed peak flow-based strategies and studies that

employed symptom-based strategies were similar with respect to

improvements in self efficacy scales and reductions in ED visits.

However, studies that employed peak flow based strategies demon-

strated greater improvement in a measure of physiological func-

tion [combined lung function (SMD 0.50, 95% CI 0.25 to 0.75)]

and greater reductions in measures of morbidity [days of school

absence (SMD -0.22, 95% CI -0.40 to -0.04), days of restricted




14/187

activity (SMD -0.68,95% CI -1.13 to -0.23), and nightsdisturbed

by asthma (SMD -0.72, 95% CI -1.09 to -0.36)], and health careutilization [general practitioner visits (SMD -0.24, 95% CI -0.55

to 0.06) and risk of hospitalization (OR 0.41, 95% CI 0.21 to

0.81; NNT 9.0, 95% CI 6.4 to 30.5)] than studies that employed

symptom basedstrategiesafter stratification and pooling. No stud-

ies reported on direct comparisons of symptom based versus peak

flow based strategies.


CARE BY INTERVENTION TYPE (N = 32 TRIALS)

When comparing studies by the type of intervention, both indi-

vidual and group interventions demonstrated similar beneficial ef-

fects on a combined measure of lung function, self-efficacy scales,

and ED visits when compared to usual care. The reductions in

morbidity measures [days of school absence (SMD -0.20, 95% CI

-0.35 to -0.04), days of restricted activity (SMD -0.68, 95% CI -

1.13 to -0.23), nights disturbed by asthma (SMD -0.86, 95% CI

-1.38 to -0.35)] and a measure of health care utilization [general

practitioner visits (SMD -0.29, 95% CI -0.56 to -0.03)] were gen-

erally stronger among trials pooled for individual interventions

as opposed to group interventions. There was a reduction in a

measure of health care utilization [hospitalizations (SMD -0.22,

95% CI -0.44 to -0.01)] that was stronger among trials pooled for

group interventions as opposed to individual interventions. How-

ever, there were no trials reporting on direct comparisons of indi-

vidual versus group interventions. Two studies employed a com-

bination of individual and group interventions (Fireman 1981;

Shields 1990), and the results from these two studies were con-

sistent with those obtained from the other studies. Overall, the

number of trials available for individual and group comparisons

of most outcomes were relatively sparse (eg between one and three

trials).


CARE BY INTERVENTION INTENSITY (N = 32 TRIALS)

When the intensity of the programs were compared, both single

and multiple session interventions were associated with similarimprovements in measures of combined lung function, exacerba-

tions, and days of school absence. However, studies employing

single sessions were associated with greater reductions in certain

morbidity measures [days of restricted activity (SMD -0.61, 95%

CI -1.12 to -0.11) and nights disturbed by asthma (SMD -0.86,

95% CI -1.38 to -0.35)] than studies employing multiple sessions.

On the other hand, studies employing multiple sessions were as-

sociated with improvement in a measure of self-perception [self-

efficacy scales (SMD 0.36, 95% CI 0.15 to 0.57)] and reductions

in measures of health care utilization [general practitioner visits

(SMD -0.17, 95% CI -0.35 to 0.00) and ED visits (SMD -0.21,

95% CI -0.33 to -0.09)] that were not seen in studies employing

single sessions. There were no direct comparisons of single ses-

sions versus multiple sessions for any reported outcome, and mostoutcomes were sparse after stratification by session number.

OUTCOMES: SELF-MANAGEMENT VS. USUAL CARE BY

TRIAL TYPE (N = 32 TRIALS)

When comparing studies by trial type, both RCTs and CCTs were

similar with regard to improvements in combined lung function

and self-efficacy scales and reductions in days of school absence.

CCTs were generally associated with greater reductions in certain

morbidity measures [exacerbations (SMD -0.47, 95% CI -0.90 to

-0.04), days of restricted activity (SMD -0.58, 95% CI -1.00 to -

0.15), and nightsdisturbed by asthma (SMD -0.86, 95% CI -1.38

to -0.35)] anda measure of healthcare utilization [hospitalizations(SMD -0.41, 95% CI -0.90 to 0.09)] than RCTs when studies

were stratified by trial type and pooled. Only for a single measure

of health care utilization [EDvisits (SMD -0.23,95% CI -0.36 to -

0.09)] were RCTs generally stronger than CCTs after stratification

and pooling of trials.

OUTCOMES: SELF-MANAGEMENT VS. USUAL CARE BY

ADEQUACY OF ALLOCATION CONCEALMENT (N = 32

TRIALS)

When studies were compared by the adequacy of allocation con-

cealment, studies judged to have both adequate and inadequate

concealment demonstrated similar improvements in combinedlung function and self-efficacy scales, and reductions in days of

school absence and ED visits. However, studies with inadequate

concealment generally had greater reductions in certain measures

of morbidity [exacerbations (SMD -0.47, 95% CI -0.90 to -0.04),

days of restricted activity (SMD -0.58,95% CI -1.00 to -0.15)and

nights disturbed by asthma (SMD -0.86, 95% CI -1.38 to -0.35)]

and a measure of health care utilization [hospitalization (SMD -

0.41, 95% CI -0.90 to 0.09)] than studies with adequate conceal-

ment when studies were stratified by the adequacy of allocation

concealment and pooled. Studies judged to have unclear conceal-

ment performedbetterthan studies in either of the other two cate-

gories for some outcome measures [combine lung function (SMD

1.24, 95% CI 0.26 to 2.22), general practitioner visits (SMD -0.72, 95% CI -1.45 to 0.01), and ED visits (SMD -0.29, 95%

CI -0.47 to -0.11)] and worse than studies in either of the other

two categories for other measures [days of school absence (SMD

-0.09, 95% CI -0.22 to 0.04), days of restricted activity (SMD -

0.12, 95% CI -0.47 to 0.24), nights disturbed by asthma (SMD

0.24, 95% CI -0.21 to 0.68), self-efficacy scales (SMD 0.23, 95%

CI -0.09 to 0.54), and hospitalizations (SMD -0.01, 95% CI -

0.16 to 0.15)]. Studies judged to have adequate concealment had

a greater decrease in a single measure of health care utilization [the

risk of hospitalization (OR 0.41, 95% CI 0.21 to 0.81; NNT 9.0,

95% CI 6.4 to 30.5)]. Generally, most outcomes were sparse after

stratification by the adequacy of allocation concealment.




15/187


CARE BY ADEQUACY OF FOLLOW-UP (N = 32 TRIALS)

When studies were compared by the adequacy of follow-up of

participants, studies judged to have adequate and inadequate fol-

low-up of participants had similar reductions in days of school

absence, days of restricted activity, self-efficacy scales, and ED vis-

its. However, studies judged to have adequate follow-up were as-

sociated with improvements in a measure of physiological func-

tioning [combined lung function (SMD 0.50, 95% CI 0.25 to

0.75)] and a measure of morbidity [nights disturbed by asthma

(SMD -0.34, 95% CI -0.62 to -0.05)] and had greater reductions

in two measures of health care utilization [risk of hospitalization

(OR 0.38, 95% CI 0.20 to 0.73; NNT 7.8, 95% CI 5.8 to 19.6)

and hospitalizations (SMD -0.25, 95% CI -0.42 to -0.07)] than

studies judged to have inadequate follow-up after stratification by

adequacy of follow-up and pooling of trials. Studies judged to

have inadequate follow-up of participants had greater reduction

in a measure of health care utilization [general practitioner visits

(SMD -0.55, 95% CI -0.99 to -0.11)] than studies judged to have

adequate follow-up after stratification and pooling. Overall, few

studies were judged to have inadequate follow-up and this created

outcomes with sparse numbers.


CARE BY ASTHMA SEVERITY (N = 32 TRIALS)

When studies were compared by the degree of asthma severity

of participants, trials that enrolled subjects with mild to moder-ate and moderate to severe asthma had similar improvements in

a measure of combined lung function and similar reductions in

exacerbations. However, studies with participants who had mod-

erate to severe asthma were associated with greater reductions in

measures of morbidity [days of school absence (SMD -0.24, 95%

CI -0.37 to -0.10), days of restricted activity (SMD-0.64, 95% CI

-1.02 to -0.25), nights disturbed by asthma (SMD -0.72, 95% CI

-1.09 to -0.36)] and health care utilization [general practitioner

visits (SMD -0.48, 95% CI -0.82 to -0.15), ED visits (SMD -

0.34, 95% CI -0.52 to -0.16), and hospitalizations (SMD -0.35,

95% CI -0.60 to -0.09)] than studies with participants with mild

to moderate asthma. However, trials stratified by asthma severity

were sparse for most outcomes, and no trials reported on direct

comparisons of asthma severity for any outcome. For a number

of studies, it was not possible to determine the asthma severity of

the participants, and the results of outcomes for these studies were

similar to those obtained for studies with participants who had

mild to moderate asthma.

D I S C U S S I O N

In this systematic review of 32 trials involving 3706 pediatric pa-

tients, asthma education programs compared to usual care were

found to improve measures of physiological function and self-effi-

cacy; reduce days of school absence and days of restricted activity,decrease emergency department utilization, and perhaps reduce

nights disturbed by asthma. However, the effects of asthma educa-

tion on the average number of nights disturbed by asthma symp-

toms was only apparent when results were pooled using a fixed-

effects model but not when a more conservative random-effects

model was employed. In addition, it was almost inevitable that

there was significant statistical heterogeneity of results for some

outcome measures (eg, nights disturbed by asthma was reported

as nights with asthma symptoms or nights of sleep interruption),

given the diversity of ways the same outcome is operationally de-

fined and measured.

Subgroup analyses were conducted to examine the impact of pa-tient or educational factors that may be important in the design

of interventions and good quality research. These results are best

viewed as tentative given the small number of pooled trials for

each subgroup post-stratification. In addition, since no studies

included direct comparisons of the different components of pa-

tient or educational factors, subgroup analyses were conducted to

compare the magnitude of effects for each subgroup. In general,

self-management education worked well for persons with moder-

ate-to-severe asthma as well as for those with mild-to-moderate

asthma, although several outcomes were stronger for those with

greater asthma severity. In looking at the characteristics of the in-

terventions, both individual and group interventions, and single

and multiple educational sessions were effective, but not neces-sarily for the same outcomes. Peak flow-based strategies generally

showed greater effect sizesthan symptom-based strategies.In look-

ing at the effects of the interventions by time, beneficial effects

on measures of physiological function were apparent within the

first six months following enrolment, but benefits did not become

fully apparent on measures of morbidity or health care utilization

until 7 to 12 months following enrolment.

Subgroup analyses were also conducted to look at the effects of

study design and methodological quality on outcomes. Both RCTs

and CCTs showed beneficial results but not necessarily for the

same outcomes. CCTs appeared to be associated with greater re-

ductions in manymeasuresof morbidityand healthcare utilizationthan RCTs. The adequacy of allocation concealment was unclear

for a large number of studies (43%), however studies judged to

have inadequate concealment performed better on many morbid-

ityand healthcare utilization measures than studies judged to have

adequate or unclear concealment. Studies with adequate follow-

up generally showed greater effects than studies with inadequate

follow-up, however studies with inadequate follow-up were few

(23%). For four outcome measures (combined lung function, days

of school absence, self-efficacy scales, and ED visits), the effects

of self-management education were consistently as beneficial or

better among pooled studies of higher quality (eg. RCTs, adequate

concealment, adequate follow-up) than among pooled studies of




16/187

lower quality.

A U T H O R S ’ C O N C L U S I O N S

Implications for practice

(1) Evidence from this systematic review of existing clinical tri-

als supports the conclusion that self-management educational in-

terventions for children and adolescents with asthma compared

to usual care result in improved physiological function, decreased

asthma morbidity, improved self-perception, and reduced health

care utilization. The educational programs included in this review

all incorporate prevention and attack management components as

part of their interventions. A small subset of the studies included a social skills component. The results suggest the desirability of in-

corporating self-management education consisting of prevention

and attack management components into routine asthma care for

children and adolescents.

(2) Evidence from this systematic review suggests that on aver-

age patients who receive self-management education would incur

moderate improvements in their physiological function over those

who receive usual care.


age patients who receive self-management education would have

modest reductions in school absences and days with restricted ac-

tivity, andmay reduce thenightsin which asthmasymptomsoccur

compared to those who receive usual care.


age patients who receive self-management education would have

moderate improvements in their feelings of self-control compared

to those who receive usual care.


age patients who receive self-management education would incur

modest decreases in their utilization of the emergency department

compared to those who receive usual care.

(6) This systematic review found no reliable differences in the

risk or frequency of hospitalizations between usual care and caresupplemented with self-management education. These types of

more rare andseriousevents maybe beyondthe ability of education

to influence.

(7) There is not enough evidence to reliably discern differences

in the effectiveness of self-management education as a function of

differences in asthma severity or the components of educational

programs. There were no direct comparisons evaluating these dif-

ferences head-to-head. However, tentative results from this review

suggest that patients with more severe asthma may derive greater

benefits from education than those with milder forms of asthma.

In addition, tentative results from this review suggest that peak

flow-based interventions may outperform symptom-based inter-

ventions. In general, benefits of education on physiological out-comes became apparent within the first six months following en-

rolment, but benefits of education on morbidity and health care

utilization measures didnot become apparent until seven to twelve

months following enrolment.

Implications for research

(1)More than half of the intervention studies identifiedwere either

poorlyreported or of less than desired quality, or both. Much more

attentionneedsto be paid to good reporting andhigh quality study

design in the future. Tentative results from this review suggest that

studies judged to be of poorer quality generally reported results

that were stronger than studies judged to be of higher quality.

(2) Many studies contained missing information on outcomes of

interest that either precluded the incorporation of their data into

pooled estimates of effect or resulted in the use of a conservative

method of imputation to obtain pooled measures of variance. Fu-

ture research efforts should assess the reliability of various methods

of imputation and encourage more complete reporting of study

results.

(3) Given that evidence supports the conclusion that education is

moreeffective thanno education, future studies shoulddirectlytest

alternative interventions against one another rather than against

no education controls.

(4) Evidence was insufficient to adequately and reliably estimateeffects for many of the important subgroups for wh

educational interventions for asthma in children

Documents