Abstract
The aim of the present study was to establish the efficacy in terms of morbidity and health-related quality of life (HRQoL) of a group asthma education-exercise programme to children with low (below 10th percentile value) quality-of-life scores.
A controlled, randomised, open, clinical trial was conducted. In total, 36 out of 53 unhappy children, among 204 (68%) respondents, treated in four paediatric practices, enrolled (mean age 10 yrs; range: 8–12 yrs), after random allocation in control and intervention groups (child, parent, teacher). Measurements were taken at baseline (T0) and after 3, 6 (T6) and 9 months (T9; intervention group only at 9 months). All but four controls completed the study.
From T0–T6, changes (Δ) in HRQoL were clinically important and significantly greater in the intervention group than in the control group, both for generic HRQoL (effect size (ES) 0.95; Δ 16%±12% versus -1±4%) and for asthma-specific HRQoL (ES 0.58; Δ 15%±17% versus 1.5±14%). T9 measurements were consistent with T6 findings. Changes in sick days (ES 0.78), oral prednisone courses (ES 0.71) and doctor visits (ES 0.74) over a 6-month period were greater in the intervention group than in the control group. Changes could not be ascribed to change in lung function or medication.
In unhappy children, quality of life and morbidity may improve with a low intensity asthma education-exercise programme, even without gains in pulmonary function or exercise tolerance.
Asthma is one of the most common chronic diseases in childhood 1, 2. With modern treatment, children with asthma in paediatric care may experience a normal life 3. Therefore, the prevalence of low well-being (quality-of-life scores below the 10th percentile value (P10)) is not expected to be elevated. In a study in Europe, half of children with asthma had persistent symptoms 4 that may lower their well-being. Well-being can be measured as generic health-related quality of life (HRQoL). The current authors’ screening of well-being in children treated in four paediatric asthma practices showed 2.5 times (25%)the normal prevalence of low well-being. As generic HRQoL is a measure of happiness, children who reported a low HRQoL level (quality-of-life score <P10) related to asthma were considered to be unhappy, and to be candidates for an intervention aimed at increasing happiness.
Improvement of generic HRQoL can be sought through better self-management of asthma. The favourable effects on morbidity of asthma educational programmes and of exercise programmes have been described previously 5–10, but no intervention study has evaluated the impact of education and exercise on quality of life in children.
A programme that includes both education and exercise has been shown to have favourable effects on HRQoL in adults 6. The current authors developed an exercise programme, superimposed on an education programme that had been validated previously 8. The purpose was to familiarise the child and parents with management of asthma at home, in school and in sports.
The objective of the present study was to establish the efficacy, in terms of morbidity and HRQoL, of this group asthma education-exercise programme, offered to children who remained unhappy despite regular asthma care offered in a paediatric practice.
The current authors’ hypothesis was that generic well-being, HRQoL and morbidity can be improved with this intervention.
METHODS
Setting and sample
This open randomised controlled trial (fig. 1⇓) was conducted in the Child Rehabilitation Centre (Beatrixoord, Haren, The Netherlands), a tertiary ambulatory clinic, between January 2004 and September 2005.
The Secondary Prevention of Asthma by Movement and Exercise (SPASME) intervention. T0: baseline; T3: 3 months; T6: 6 months; T9: 9 months. ––––: comparisons between control and intervention groups. - - - -: longitudinal comparison.
The accessible population was defined as any child 7–12 yrs old with asthma visiting four paediatric asthma clinics in the region. The parents of all subjects were sent a letter introducing the screening of HRQoL 11–15, including a preaddressed and stamped reply envelope.
The response rate was 68%. Children who had a generic HRQoL score <P10 in the self-report form (26%) were selected. The parents of these children were telephoned to verify that low well-being was not caused by a recent asthma exacerbation, another chronic disorder or any cause of unhappiness. Parents were sent a letter introducing the study and an invitation to participate in the study.
Participants were assigned randomly to a control group and an intervention group. Both groups continued regular care and visits to the referring paediatrician and asthma nurse. In addition, the intervention group received a 3-month training programme. The intervention tested was group education-exercise for the child, and education to caregivers and schoolteachers.
Inclusion criteria for the current study were: total generic HRQoL score <P10, age 7–12 yrs, diagnosed asthma based on international criteria (American Thoracic Society 16), use of asthma medication according to guidelines, no recent asthma diagnosis, exacerbation or infection.
Exclusion criteria were other causes of low HRQoL and cognitive level <7 yrs.
Data collection
The paediatrician and the asthma nurse at the centre took the history, and performed a physical examination and inhalation test at baseline (T0), after the 3 month intervention or control period (T3) and 3 months after discharge (T6). Medication was adjusted at T0, if judged necessary by the paediatrician for practical reasons (better timing), or for newly found exercise intolerance (measured using the exercise-induced bronchoconstriction (EIB) test) 17. Adjustments were made according to guidelines from the Dutch Paediatric Respiratory Society 18.
As a measure of lung function, forced expiratory volume in one second (FEV1; % predicted) and the FEV1/forced vital capacity (FVC), performed at T0 and T3. Exercise intolerance was determined using the EIB test (fall in FEV1 during the test, cut-off 10%) 17.
Exercise tolerance was related to distance run in the EIB test and a 6-min running test indoors.
Morbidity parameters over the previous 6-month period consisted of: asthma severity score (on a Likert scale from low (1) to severe (7)); number of visits to the general physician, emergency room and hospital; number of asthma attacks and of oral steroid courses; and days absent from school (child) or work (parent), as reported by parents.
Demographic data included sex, age, weight, length of the child, parental age, educational level and employment status of the parents.
HRQoL measures were taken at T0, T3 and T6 and 6 months after discharge (T9). Only the intervention group completed the T9 measurement.
The Dutch Children Academisch Ziekenhuis Leiden(AZL; Academic Hospital Leiden, now called University Medical Centre Leiden)–Toegepast-Natuurwetenschappelijk Onderzoek (TNO; National Research Institute for Prevention and Health) Quality of Life questionnaire (DUCATQOL; also called DUX-25) 11, and the TNO–AZL Children Quality of Life questionnaire (TACQOL; both the generic 12–14 and asthma-specific 15 modules) were used. The DUX-25 11 collects information in four domains (physical, at-home functioning, emotional and social). The DUX-25 score has discriminative properties (Crohnbach alpha 0.94) and was used to recognise children with a level of well-being below the P10 of the reference population (DUX-25 total score <63). Children with a self-reported DUX-25-score <P10 were considered unhappy.
The TACQOL generic questionnaire 12–14 evaluates five domains (physical, motor, autonomic, school and social functioning) and two moods (positive and negative).
The TACQOL asthma questionnaire 15 evaluates five domains (complaints, emotions, situations, treatment and medication demands). For the purpose of the present study, a TACQOL total score (summation of domain scores) was calculated.
In the two TACQOL questionnaires 12–15, HRQoL scores are influenced by both the presence and the perception (the emotional response) of having a health problem. Responses are combined into a single score (fig. 2⇓), that can either be favourable, in the absence of problems (4) and in the presence of health problems that are “fine” (3), or unfavourable in the presence of problems that are perceived as “(quite) bad” (1–0).
The scoring system for the Toegepast-Natuurwetenschappelijk Onderzoek–Academisch Ziekenhuis Leiden Children Quality of Life (TACQOL-asthma) questionnaire.
Raw quality-of-life scores were adapted to represent percentages of the maximum possible DUX-25 and TACQOL domain scores. Higher scores indicate better HRQoL, in both questionnaires.
The intervention
The education-exercise programme was based on the view 9, 10, 19 that education by play and exercise are natural modes of communication for pre-adolescent children 9, 19. As group education may be more effective in children 9, 19, the education-exercise programme was offered to groups of 8-10 children, in ten weekly sessions of 1.5 h of education and 1 h of exercise, and in a follow-up session at T6, 3 months after discharge. Parents followed five educational sessions (duration 1.5 h), every 2 weeks. Teachers attended one 1.5-h session.
The programme was offered by a paediatric nurse and child physiotherapist trained in the original educational programme by Colland 8. This programme has been evaluated and validated 8 previously, and uses games and learning materials designed specifically for the school age group (7–12 yrs). Educational themes were offered in a 2-way dialogue setting, as described in the original study 8. Both group education and individual advice and reassurance 17 on capability to perform activities in daily life and sports were offered.
A physical exercise programme was added using elements of other exercise programmes, as reviewed by Veldhoven 10. Contrary to other programmes, the current authors’ exercise programme had no focus on improving exercise tolerance, for which a higher frequency (three times per week) of training is needed.
Written informed consent for the study was obtained from the parents of all responding children, as well as from 12-yr-old children themselves.
The Medical Ethical Board of the University Medical Centre Groningen (Groningen, The Netherlands) and the participating referral hospitals approved the study.
Analysis
The sample size was calculated using a sample size analysis to determine the number of children needed for the study effect 20. A difference of 1 sd, a power of 80% and a significance level of 5% was used to calculate sample size. Based on these assumptions, the number of patients required was 32.
Allocation to the intervention group or the control group was done using the minimisation procedure by age, sex and lung function (FEV1 >85% pred, or FEV1 ≤85% pred) 20. This procedure is an adequate method for blind group allocation in small samples. The principle of analysis by intention to treat was followed.
Changes in HRQoL scores were calculated between T0 and T3 and between T0 and T6.
Changes in morbidity were calculated over the 6-month periods prior to and following T0, but data were corrected to a 1-yr period.
Comparison between groups on HRQoL changes and morbidity changes was performed using the Mann–Whitney test.
Within the intervention group, morbidity and quality-of-life scores before and after T0 were analysed pairwise, using the Wilcoxon test 20.
Adjustment for age, sex, educational level of caregivers and severity of asthma was not performed, as no differences between groups were found.
The recommendations by Middel et al. 21 were followed to study the magnitude (effect size) of HRQoL changes between T0–T3, T0–T6 or T0–T9 21. Changes with an effect size of 0.50 or more were defined as “clinically important” 21.
For all tests, a level of significance of p<0.05, two-sided, was used. Data are presented as mean±sd, unless otherwise stated.
RESULTS
Out of 301 children approached (fig. 3⇓), 204 (67.8%) provided complete responses. A well-being level of >P10 was seen in 74%, and of <P10 in 26%. Differences in well-being levels between the children with scores below or above the cut-off score were not explained by differences in morbidity parameters or in lung function values.
Flow diagram showing progress through the phases of the study. P10: 10th percentile value.
Out of the 53 unhappy children invited to participate in the study, 40 enrolled.
Asthma severity, medication used and prevalence of low DUX-25 total score did not differ significantly between participants from the different hospitals.
Parents gave a mean±sd asthma severity score of 4.5±2.8 on a Likert scale of 1–7 (1 not severe, 7 severe). The paediatrician classified children's asthma as moderate-to-severe, and persistent 18. They had been treated according to guidelines 18. Lung function values were in the normal range (FEV1 and FEV1/FVC 4.6% and 3.5%, respectively, below predicted values).
The mean±sd age was 10±1 yrs, length 147±3 cm and weight 40±2 kg. Overall, 53% were male. Parental age was 35±3 yrs. At least one parent was employed in 92% of cases, while 8% were dependent on social security. A total of 30% had received previous education sessions. As expected, DUX-25 total score was significantly lower (56.9±11.5) than the norm (78±13; p<0.001).
At T0, two children in each group had to be excluded: two had recently acquired another chronic disorder prevailing over asthma, and two had a recent exacerbation. The 36 remaining children were divided between the control and intervention groups after minimisation 20.
Comparisons between intervention and control groups
At T0, pulmonary function did not differ between groups, neither in FEV1 (control group 95.6±13.5% pred; intervention group 98.6±11.4% pred), FEV1/FVC (control group 84.5±8%; intervention group 84.5±9%). Over the following 3 months, pulmonary function did not improve significantly in either groups.
T0 running distance did not differ between groups, and did not change between T0 and T3. It was measured in the EIB test (control group 750±158 m; intervention group 744±232 m), and in the 6-min indoor running test (1000±400 m in both groups).
In three new EIB cases, medication adaptation led to an improvement in FEV1 fall in the EIB test, consistent with disappearance of secondary exercise intolerance (T0: fall 15±11%; T3: fall 6.7±13%).
Optimisation of medication was applied for better timing of medication in four children, and for newly diagnosed EIB in six cases, spread between the intervention and control groups. Comparison between those without and those with medication alteration, on changes in morbidity and HRQoL (T0–T6), showed no differences, but the groups were too small to draw reliable conclusions.
At T0, morbidity parameters between the intervention and control group did not differ (table 1⇓); nor did asthma severity score (control group 4.2±2.7; intervention group 4.9±2.8 on a Likert scale ranging 1–7).
Morbidity at baseline(T0) and changes over a 6-month period in the intervention and control groups
Comparison between groups on changes in morbidity parameters over the 6-month period between T0–T6 (table 1⇑) showed greater changes in the intervention than in the control group for visits to the general physician and paediatrician (effect size 0.71), oral steroid (prednisone) courses (effect size 0.71), and days absent from school (effect size 0.74).
HRQoL scores at T0 did not differ between the intervention and control groups.
Comparison of changes in HRQoL over the 3-month period between T0–T3 (table 2⇓), showed greater changes in HRQoL in the intervention group than in the control group, for two self-reported total scores (DUX-25 and TACQOL-asthma) and for four proxy-reported domains (TACQOL-asthma situations and TACQOL-generic motor, autonomic and cognitive domains).
Changes in quality of life over a 3-month period from baseline in the intervention and control groups
Comparison of changes in HRQoL over the 6-month period between T0–T6 (table 3⇓), showed greater changes in HRQoL in the intervention-group than in the control-group, for almost all self and proxy reported HRQoL-domains.
Changes in quality-of-life over a 6-month period from baseline in the intervention and control groups
Comparison of changes over time between T0–T6 within the intervention group (table 4⇓) showed changes with effect size greater than 0.05 sd that were of clinical importance, for all total HRQoL scores, in eight out of 16 child-reported, and 12 out of 16 parent-reported domains. Changes in HRQoL scores over a longer period (T0–T9) were consistent with findings at T6, but were available in only 9 out of 18 children. No reliable conclusions could be drawn.
Self and proxy report on health-related quality of life (HRQoL) at baseline (T0), and changes over 6 months (ΔT0–T6) in the intervention group
During the study period, the drop-out percentage was surprisingly low. All children in the intervention group, and 14 children in the control group completed 6 months’ follow-up measurements. Attendance in the intervention group was 84.5% for children and 79.4% for parents, absence being due to sick leave for viral infections. Attendance by teachers was 42%.
DISCUSSION
As far as the current authors are aware, the present study is the first paediatric study to show improvements in quality of life following a 3-month asthma education-exercise programme. In school-age children with low scores in an initial generic HRQoL self-report, clear improvements in HRQoL and reductions in oral steroid courses, school absence and doctor visits are seen following the intervention. These may not be mediated simply by changes in medication control or in clinical parameters.
The effects of asthma educational programmes on morbidity in children 5, 7–9, evaluated in a recent meta-analysis 5, showed reduced morbidity, emergency room visits and days absent from school. Using the educational programme of Colland 8, the present authors replicated reductions of paediatric visits and school absenteeism as shown in other studies 5, 8–9.
The effects on HRQoL of a combination of education and exercise have not been studied in children. In accordance with an adult education-exercise program 6, the present authors showed that HRQoL, both asthma-specific and generic, can improve significantly, even when gains in pulmonary function or exercise tolerance are not demonstrated.
The effects seen immediately after the current programme were smaller than those 3 or 6 months later. These late effects may be understandable from the nature of HRQoL, which evaluates functioning in physical, personal and social domains of daily life. It is likely that the process of self-management takes time to accomplish effects in all aspects of daily life, and therefore HRQoL effects may need time to develop.
The present authors found one randomised controlled trial of group education in a large population of children with asthma that did show effects on morbidity, but not on disease-specific HRQoL 22. Children in the study had a relatively high HRQoL at the start of the study, and they received three sessions (45–60 min) of education, as opposed to the present unhappy group, which received 10 sessions 22.
Two studies on disease-specific-HRQoL after the start of asthma medication in medication-naive children were found 7, 23. The changes (effect size) in TACQOL-asthma scores in the current education-exercise group were similar to the changes in the one study 7, but were slightly smaller than changes in the other 23. Changes in these studies were greater than the minimal important change for the Paediatric Asthma Quality-of-Life Questionnaire 24.
The HRQoL changes seen in the present study also reach the level of clinical importance (effect sizes >0.50). What is new is the finding that clinical important changes can be found after education-exercise in children taking asthma medication. It is also new that effects were shown in the self and proxy reports of generic and asthma-specific measures.
It remains uncertain which component of the present training programme may have contributed to the positive effects on quality-of-life, as differential effects were not studied 25. However, the educational programme used 8 has proven effects on anxiety and coping, due to better self-management 8. As the current authors were trained in this educational programme 8, it was hypothesised that improved coping and anxiety, though not measured, may have been a component that contributed to improved HRQoL.
Improved subjective feelings may have been another component contributing to the observed HRQoL improvements. The TACQOL(-asthma) measures used incorporate both the number and perception of health problems in the scoring (fig. 2⇑). This makes the TACQOL sensitive to changes that are consistent with improved perceptions and qualities of health problems, even in the face of unchanged quantities.
It is unlikely that medication change or better inhalation technique had sufficient effect on HRQoL changes. The comparison between children with and without alteration of medication showed similar HRQoL improvements in both groups.
Other methodological considerations
The present findings are applicable only in children aged 7–12 yrs who suffer a substantial loss of well-being from persistent asthma symptoms, despite paediatric care.
Although the current study included a small number of children, this was sufficient according to the power analysis carried out in advance. The population was selected, as children with low HRQoL scores (pre-defined self-reported DUX-25 total score <P10) were eligible to enrol, and they represented a quarter of children with asthma. Willingness to participate was greater (75%) than the one-third expected from other studies [26, 27], and attendance by children and parents was high. The minimisation procedure was used to reduce the effects of selection on the findings. It was assumed that the low drop-out percentage prevented further selection bias.
The present findings need to be reproduced in children with asthma and a normal HRQoL. Further study is needed to draw conclusions about long-term effects 26.
The message of the present study would be that, in addition to monitoring pulmonary function and disease-specific status in childhood asthma, paediatricians must monitor generic well-being, as a screening measure to recognise “unhappy” children. Happiness can be gained from an education-exercise programme that may enable these children to achieve better self-management of asthma, leading to less morbidity and greater well-being.
Support statement
The present study was supported by de Vereniging Beatrix Noord Nederland (Haren, The Netherlands).
Statement of interest
None declared.
Acknowledgments
The current authors would like to thank the team of the Dept of Child Rehabilitation (Centre for Rehabilitation, University Medical Centre Groningen, Groningen, The Netherlands). C. van Baak, E. Beens, A. Swart, M. Eskes and S. Pariger, for the development and teaching of the training programme, and H. Eeftingh for the distribution and collection of questionnaires, as well as the parents and children for their participation in the programme and in the measurements.
- Received October 24, 2007.
- Accepted June 30, 2008.
- © ERS Journals Ltd
References
