Abstract
The costs and effectiveness of asthma action plans for children were evaluated in a cross-sectional economic analysis. Direct health care and indirect costs, nights with symptoms, and asthma attacks were measured in 879 Ontario children with asthma. From a societal perspective, the total annual costs of the asthma action plan and the control groups were CDN$6, 948 and CDN$6, 140 per patient, respectively. Health outcomes were similar. The difference in cost was attributable to greater medication and health services use in the intervention group. Prospective randomized trials are necessary to measure potential improvements in control of asthma using asthma action plans.
Keywords: action plans, asthma, child, cost-effectiveness, disease management, health services utilization
Introduction
In Canada, the prevalence of asthma in children under the age of 14 years is 11% (1). Asthma is the leading cause of school absences in Canada, and evidence suggests that asthma care for children is inadequate, resulting in less than optimal control (2, 3). In the last two decades, asthma has become a major cause of increased morbidity and mortality in children in various global regions due to the under-treatment of the disease (4, 5). It has been hypothesized that much of the burden of pediatric asthma morbidity may be prevented with proper asthma education and disease management.
In addition to the burden of morbidity due to disease, asthma poses a significant economic burden. Direct and indirect expenditures for care of children with asthma in the U.S. in 1996 were estimated to be close to US$2 billion, with direct medical expenditures related to prescription medication, hospital inpatient stay, hospital outpatient care, emergency department (ED) visits, and office-based visits accounting for approximately half (6). Parents’ loss of productivity from asthma-related school absence days was US$719 million, and 211 school-age children died from asthma in 1996, representing US$265 million in lost potential lifetime earnings (6).
Asthma education and action plans are recognized as important elements for the optimal management of asthma and to avoid an acute asthma attack in adults, but evidence is inconclusive regarding their effectiveness and cost-effectiveness in children (7). Studies have demonstrated the short-term benefits of asthma disease management programs, such as improved outcomes, efficient use of resources and decreased costs (8, 9). However, evidence on their long-term benefits is lacking. This evidence is critical since the severity of asthma in a patient may vary from year to year (8). A Cochrane Systematic Review (2004) on educational interventions for asthma in children stated that limited evidence existed to conclude whether self-management education is effective for children with more severe asthma compared with those with a less severe form of the disease (9).
What constitutes asthma education for children may vary from sessions with a qualified asthma educator to simple written materials and resources. Health care providers often consider a written asthma action plan accompanied by education sessions as the gold-standard in asthma self-management. An asthma action plan typically consists of written instructions that are customized for the patient. The plan includes (1) a list of the patient’s asthma triggers and how they should be avoided; (2) a list of symptoms to be aware of and actions to take should they occur; and (3) the names and doses of medications that the patient requires and when to use them. The objective is to regularly update the plan and follow it closely to prevent or effectively treat asthma symptoms and exacerbations (10).
Previous research has suggested that the success of an asthma action plan in preventing exacerbation and for day-today asthma control may depend on factors such as age, gender, ethnicity, parental education, immigrant status, language, and financial constraints on the family (11–13). The main reasons cited for failure of an asthma action plan implementation are insufficient time provided by physicians to explain the plan to caregivers or inadequate monitoring by physicians (14, 15). If effective, asthma action plans may be expected to increase costs related to medication use and regular physician follow-up but reduce costly inpatient admissions and emergency room visits. Although three economic evaluations on education and self-management programs related to pediatric asthma were identified, none explicitly measured the cost-effectiveness of asthma action plans (16). The primary research objective of this study was to evaluate the economic benefit of asthma action plans for children by comparing the direct and indirect costs and health outcomes between children receiving an asthma action plan and those without such a plan.
Methods
Study Population and Data Collection
A stratified sampling plan was used to recruit 879 children with asthma from seven urban and suburban sites in the greater Toronto area, including primary care practices, respiratory specialist offices, asthma clinics, and EDs. While the sampling was not random, the stratified design allowed for recruitment of children with a wide spectrum of asthma severity. Children who were between 1 and 18 years of age with a charted clinical diagnosis of asthma or reactive airway disease and a prescription for at least one asthma medication, such as inhaled corticosteroids or bronchodilators, within the past year were eligible. Detailed information regarding family demographics, asthma history, asthma medication use, asthma symptoms and exacerbations, use of health services, and receipt of asthma education and actions plans was collected by personal interviews with older children and with the parents of infants and young children between November 2000 and March 2003. Complete data were available from 217 subjects who responded affirmatively to the question, “Has your child ever been given a personal asthma self-management or action plan?” (intervention group) and from 618 subjects who responded negatively (control group). The direct and indirect costs, the number of nights with symptoms, and the number of asthma attacks were compared between the intervention and control groups. The study was approved by the Hospital for Sick Children and the University of Toronto Research Ethics Boards, and all participants provided written informed consent.
Asthma Action Plan Intervention
In addition to determining whether participants received an asthma action plan, all subjects were asked whether they received any asthma information or education in the previous 6 months. Those who responded affirmatively reported receiving information from various sources, including physicians, asthma educators or nurses, pharmacists, respiratory therapists, community organizations, or others. As asthma action plans are ideally provided alongside asthma education, those who indicated receiving an asthma action plan were assumed to have also received asthma education that included two information sessions given by an asthma educator and written materials. According to the Canadian Asthma Consensus Guidelines, patient education and asthma action plan implementation are two basic components of asthma management (17).
Cost Measurement
Respiratory-related direct and indirect costs were assessed from a societal perspective to provide the most comprehensive assessment of costs associated with asthma management. Societal costs included all items in the public and private health care sectors and the indirect productivity cost associated with caregiver time losses due to caring for children with asthma. The time horizon for the analysis was one year to facilitate comparison with other studies.
The cost of each item or service was calculated by multiplying the volume of use by its unit price. Direct health care costs consisted of asthma-related health care resource use including ambulatory visits to family doctors, pediatricians and specialists, ED visits, hospital admissions, prescriptions for asthma medications, and medical equipment (e.g., nebulizer, spacer, and peak flow meter). Health care utilization was recalled over periods that have been documented to be reliable (18). Thus, uncommon events such as ED visits and hospital admissions were recalled over one year while shorter recall intervals were used for more frequent events, such as physician visits (6 months) and use of medications (current). All health care resource use was extrapolated to one year for the analysis. Indirect costs consisted of the productivity losses of parents or caregivers to provide care related to their child’s asthma illness, including travel and wait time associated with accessing respiratory health care services. Unit prices for all items were derived from the provincial physician fee schedule, provincial drug formulary, inpatient case costing database (19), Statistics Canada wage database (20), and self-reported responses. Table 1 lists the direct and indirect cost items and the unit price sources. All prices are reported in 2003 Canadian dollars. Some unit prices for earlier years were adjusted to 2003 dollars using the consumer price index provided by the Bank of Canada.
Table 1.
Direct and indirect cost item prices and sources.
| Item | Unit price* | Cost source | |
|---|---|---|---|
| Emergency department physician consult | 80.75 | Ontario Schedule of Benefits & Fees | |
| Primary care physician visit | 29.95 | Ontario Schedule of Benefits & Fees | |
| Respiratory specialist visit | 57.10 | Ontario Schedule of Benefits & Fees | |
| Pediatrician visit | 53.15 | Ontario Schedule of Benefits & Fees | |
| Family practice visit | 54.10 | Ontario Schedule of Benefits & Fees | |
| In-patient physician assessment | 125.00 first day; 23.00/day for remaining days | Ontario Schedule of Benefits & Fees | |
| Emergency department visit | 141.21 | Chedoke-Mc Master Hospital | |
| Asthma in-patient case cost | 1 to ≤4 years | 836.90 | Ontario Case Costing Initiative |
| >4 to ≤11 years | 860.60 | ||
| >11 to ≤18 years | 803.40 | ||
| Dispensing fee—public plan | 6.54 | Ontario College of Pharmacists | |
| Dispensing fee—private plan | 11.99 | Community chain pharmacy | |
| Asthma prescription | 4.64–204.47 | Ontario Drug Benefits Plan | |
| Asthma educator | 26.00 | Asthma Clinic at the Hospital for Sick Children | |
| Asthma education brochures | 5.00 | Ontario Lung Association | |
| Asthma action plan | 0.25 | Ontario Lung Association | |
| Productivity time loss (hourly wage=) | 0–31.76 | Chedoke-Mc Master Hospital | |
| Spacer | 19.99 | Community chain pharmacy | |
| Nebulizer | 129.99 | Community chain pharmacy | |
| Peak flow meter | 43.00 | Community chain pharmacy | |
| Homemaker time loss (hourly wage=) | 9.13 | Statistics Canada | |
All prices reported in 2003 Canadian dollars.
= Hourly wage based on a 37.5–hour work week.
An inpatient case cost was retrieved from the Ontario Case Costing Initiative (OCCI) using ICD10 code J45 for asthma as the most responsible diagnosis during the admission (21). For each inpatient admission, the OCCI average daily cost for Ontario children was multiplied by the average length of stay (LOS). The physician cost for inpatient care consisted of one full consultation assessment fee with a subsequent general assessment fee assigned to each remaining day of the LOS. Medication costs were based on the Ontario Drug Benefit Program formulary prices in 2002. Asthma medication is typically prescribed in 1-month supply inhalers. As some medications were used intermittently and to allow for less than optimal compliance, it was conservatively assumed that each drug prescription reported by study participants would last 6 weeks. The total annual medication cost per child was estimated by multiplying the cost of each prescription by eight for an annual cost and summing all medication costs.
Intervention Cost
The model used to estimate the unit cost of an asthma action plan assumed two information sessions with an Asthma Educator and written materials. A certified Asthma Educator/Nurse Practitioner usually conducts two sessions of 10 to 15 minutes each. Based on nurses’ hourly wages and associated benefits at the Hospital for Sick Children in Toronto, the total cost of education was estimated to be $26.00. Written materials provided to the individual patient or caregiver consisted of booklets published by the Ontario Lung Association. The value of reading materials was estimated to be $5.00 and the cost of the printed asthma action plan itself was only $0.25 per copy. The total assigned cost for the intervention was $31.00 per patient.
While children in the control group did not receive an asthma action plan, many reported receiving other forms of asthma education. Different unit costs were assigned for asthma education based on the reported source of education. If parents reported receiving education from respiratory specialists, asthma educators, and respiratory therapists, they were allotted the full intervention cost even if they did not receive an action plan. If they reported receiving information from family physicians, pediatricians, school or community health nurses, pharmacists, community organizations, teachers, or the Internet, they were assigned a partial asthma education cost that consisted only of the cost of written education material ($5.00). Thus, the control group included participants who reported that they did not receive an asthma action plan but who may have received partial or full asthma education.
Indirect Costs
As parental productivity losses were not ascertained directly, the volumes of time losses were estimated based on the frequency of various types of health care visits. For the productivity cost for the parent or caregiver of children with asthma, we assumed a school day loss was equivalent to the productivity loss of one day of work by one parent (22). Based on Krahn et al.’s study, we assumed that one third of outpatient physician and ED visits occurred during work hours (23). An ED visit was conservatively assumed to be 3 hours (23). Family physician and pediatrician visits were assumed to last one hour, whereas specialist visits were assumed to be 1.5 hours to account for extra travel time required to visit a specialist.
A daily wage was calculated by dividing the reported annual income by 239 total working days per year. This number was derived by subtracting weekend days (104), statutory holidays (12), and vacation days (10) from 365 days. An hourly wage was derived by dividing this value by 8 hours. For parents who were students, homemakers, on social assistance, or on disability, 100% of the average Ontario homemaker’s 2001 salary reported in the CENSUS was used and adjusted for inflation to 2003 (20). The total time loss in hours associated with each child was multiplied by the hourly wage.
Outcome Measurement
Clinical guidelines state that night-time awakening due to asthma symptoms and frequency of asthma exacerbations are both important measures of asthma control in children (17). In this study, the two outcome measures were the number of nights with symptoms in the prior month and the number of attacks in the prior 6 months. A night-time symptom was defined as any wheezing or whistling in the chest, shortness of breath, chest tightness, and cough that awakened the child from his or her sleep. An attack was defined as a sudden worsening of symptoms that required action such as taking additional medicine to relieve symptoms or an unscheduled visit to an ED or doctor. The frequency reported for each outcome measure was linearly extrapolated to determine the number of annual occurrences.
Statistical Analysis
A cost-consequence analysis was used to measure the total costs and the health outcomes associated with the asthma action plan and control groups. The mean cost per patient per year and the mean number of nights with symptoms and number of attacks per patient per year were calculated and reported separately for each group. Initially, costs and outcome variables were compared using parametric and non-parametric univariate statistics. The chi-square test was applied to categorical variables and the Wilcoxon rank sum test was used for continuous variables. Incremental mean annual costs per patient and increments in the mean annual number of nights with symptoms and the number of attacks per patient were calculated between the intervention and control groups. Health service utilization was compared between groups using a chi-square test of proportions. For each variable in this category, the mean number of annual visits per patient was calculated.
As total annual cost per patient was the major dependent variable, a generalized linear model multiple regression was performed to explain the effect of independent variables on the total cost per patient and to control for potential selection bias due to the nonrandomized design. In addition to the binary exposure variable of receipt of an asthma action plan, explanatory variables included in the regression model were ethnicity, income, employment status, asthma duration, season of study participation, and number of nights with symptoms. An important potential confounder, disease severity was incorporated into the regression model using the previous use of oral systemic steroids (prednisone) as a surrogate marker for more severe disease. Because the cost data were positively skewed, a log transformation for each patient’s cost was conducted to normalize the data. Most cost predictors in the regression model were statistically significant at the 0.05 level. Total costs were retransformed using a smearing estimator to ensure that the retransformed costs were unbiased and consistent (24).
Sensitivity Analysis
Three one-way sensitivity analyses were performed to test the robustness of our findings to variations in underlying assumptions. The selection of these three analyses was based on assumptions made regarding the average daily cost of in-patient care and the productivity time loss in terms of hours. The base-case inpatient case cost assumed an average daily cost for hospital admission multiplied by the average length of stay. Physicians’ fees were added. In the first sensitivity analysis, the inpatient case costs were varied in accordance with the minimum and maximum inpatient daily cost. In the second sensitivity analysis, alternative durations of parental time losses for physician, specialist, and ED visits were tested (23). This included a sensitivity analysis that assumed a constant time loss of 1.5 hours for all types of physician and ED visits. The final sensitivity analysis assumed that all physician and ED visits occurred entirely during working hours rather than one third of working hours.
Results
Sample Characteristics
Table 2 presents the sample characteristics by action plan status. The intervention group was on average one year older than the control group. As well, a significantly greater proportion of parent respondents in the action plan group were either born in Canada or were of Canadian/North American background or spoke English at home compared to the control group. A significantly higher proportion of children in the action plan group displayed a history of allergies. In the control group, 15% received asthma education in the last 6 months from sources comparable to the intervention group, 22% received written materials only, and 63% of the control group did not receive any asthma education. There were no differences in household income, parent education, or marital status between groups.
Table 2.
Sample characteristics by group.
| Action plan n = 217
|
No action plan n = 618
|
||||
|---|---|---|---|---|---|
| Characteristics | n | % | n | % | p value* |
| Child’s sex | |||||
| Male | 132 | 60.8 | 378 | 61.2 | 0.93 |
| Female | 85 | 39.2 | 240 | 38.8 | |
| Child’s age* | |||||
| 4 years and under | 53 | 24.4 | 217 | 35.1 | 0.001 |
| 5 to 9 years | 94 | 43.3 | 269 | 43.5 | |
| 10 years or more | 70 | 32.3 | 132 | 21.4 | |
| Mean age, standard deviation | 7.73, 4.29 | 6.57, 4.10 | |||
| Mean household income, standard deviation | $63,804, $23,683 | $61,903, $24,467 | |||
| Parent respondent born in Canada* | |||||
| Yes | 146 | 62.8 | 338 | 54.6 | 0.001 |
| No | 71 | 32.7 | 280 | 45.3 | |
| Family ethnicity* | |||||
| Canadian/North American | 125 | 57.6 | 276 | 44.7 | 0.004 |
| European/East European | 40 | 18.4 | 135 | 21.8 | |
| Other | 52 | 24.0 | 207 | 33.5 | |
| Drug plan | |||||
| Yes | 189 | 87.1 | 533 | 86.3 | 0.75 |
| No | 28 | 12.9 | 85 | 13.8 | |
| Child history of allergies* | |||||
| Yes | 144 | 66.4 | 326 | 52.8 | <0.001 |
| No | 73 | 33.6 | 292 | 47.2 | |
Chi-square p value ≤0.005.
Health service utilization by subgroups of patients with one or more visits is displayed in Table 3. A significantly higher proportion of the action plan group was followed by a respiratory specialist compared with the control group. In addition, the intervention group had, on average, fewer routine family doctor visits. Although not statistically significant, lower proportions of the action plan group had experienced all three measures of urgent care, including asthma ED visits, hospital admissions, and unscheduled clinic visits for asthma, compared to the no asthma action plan group. However, among the subgroup of children who did require urgent care, the action plan group demonstrated higher rates of use, suggesting that they may represent a population with more severe asthma or asthma that is more difficult to treat.
Table 3.
Health services utilization by group for patients reporting one or more visits.
| Action plan n = 217
|
No action plan n = 618
|
|||
|---|---|---|---|---|
| Asthma-related health service | n | % | n | % |
| Family physician visits* | 91 | 42.0 | 334 | 54.1 |
| Mean visits per year; median; interquartile range | 1.31; 0.00; 2.00 | 1.73; 1.00; 2.00 | ||
| Pediatrician visits | 57 | 26.3 | 192 | 31.1 |
| Mean visits per year; median; interquartile range | 0.91; 0.00; 1.00 | 0.87; 0.00; 1.00 | ||
| Respiratory specialist visits* | 173 | 79.7 | 329 | 53.2 |
| Mean visits per year; median; interquartile range | 2.09; 2.00; 2.00 | 1.16; 1.00; 2.00 | ||
| Hospital admissions | 51 | 23.5 | 169 | 27.4 |
| Mean per year; median; interquartile range | 0.49; 0.00; 0.00 | 0.43; 0.00; 1.00 | ||
| Emergency department visits | 111 | 51.1 | 351 | 56.8 |
| Mean visits per year; median; interquartile range | 1.44; 1.00; 2.00 | 1.29; 1.00; 2.00 | ||
Chi-square p value <0.01.
Comparison of Costs
Table 4 reports the unadjusted mean direct and indirect costs for the two study groups. The asthma action plan group had a total annual cost of $6,948 per patient while costs for the control group were $6,145 per patient. Direct costs accounted for 37% and 36% of the total for the intervention and control groups, respectively, while indirect costs due to parental productivity losses accounted for 63% and 64%, respectively. The largest direct cost component was inpatient costs which accounted for 13% and 14% of the total cost for the intervention and control groups, respectively. The asthma action plan group incurred higher costs for respiratory specialists. There was also a difference in the asthma medication costs (including dispensing fees) between the two groups, with the intervention group spending an average of $166 more per child per year.
Table 4.
Unadjusted annual cost per patient by group.
| Action plan n = 217
|
No action plan n = 618
|
|||
|---|---|---|---|---|
| Cost category | Average annual cost per patient ($) | % of Total | Average annual cost per patient ($) | % of Total |
| Direct costs | ||||
| Inpatient care | 937 | 13.5 | 832 | 13.6 |
| Emergency visits | 320 | 4.6 | 286 | 4.7 |
| Family physician services | 142 | 2.1 | 188 | 3.1 |
| Respiratory specialist services | 239 | 3.4 | 133 | 2.2 |
| Pediatrician services | 97 | 1.4 | 92 | 1.5 |
| Asthma medication costs | 505 | 7.3 | 374 | 6.1 |
| Dispensing fees | 272 | 3.9 | 238 | 3.9 |
| Nebulizers | 35 | 0.5 | 38 | 0.6 |
| Spacers | 13 | 0.2 | 13 | 0.2 |
| Peak flow meters | 12 | 0.2 | 5 | 0.1 |
| Asthma education | 27 | 0.4 | 6 | 0.1 |
| Subtotal direct costs | 2,599 | 37.4 | 2,205 | 35.8 |
| Indirect costs | ||||
| Parent’s productivity loss | 4,350 | 62.6 | 3,940 | 64.2 |
| Total unadjusted costs | 6,948 | 6,145 | ||
All costs reported in 2003 Canadian dollars.
Table 5 presents the regression analysis for the adjusted total cost per patient. Costs were controlled for asthma severity (previous use of an oral steroid), frequency of night-time symptoms, and demographics characteristics, such as ethnicity, income, and employment status. The results demonstrated that higher total costs per patient were associated with more severe disease, higher household income, more nights with symptoms, and longer asthma duration. Children of parents of European or other non-North American ethnicity or who were employed also demonstrated higher costs. In addition, higher costs were associated with intermediate or peak asthma season.
Table 5.
Regression analysis of log of total cost per patient.
| Independent variable | Regression coefficient | Standard error | p value |
|---|---|---|---|
| Asthma action plan—No (reference: yes) | −0.0991 | 0.0834 | 0.24 |
| Oral steroids—no (reference: yes) | −0.7756 | 0.0864 | <0.0001 |
| Ethnicity(reference: Canadian/North American) | |||
| European/East European | 0.0143 | 0.0949 | 0.88 |
| Other | 0.3040 | 0.0867 | <0.01 |
| Income | 0.0422 | 0.0142 | <0.01 |
| Asthma duration | 0.5303 | 0.1764 | <0.01 |
| Employment (reference: full- or part-time) | |||
| Social Assistance or Disability | −0.3113 | 0.1804 | 0.085 |
| Homemaker or unemployed | −0.3899 | 0.0911 | <.0001 |
| Allergy season of participation (reference: low season) | |||
| Intermediate | 0.3031 | 0.1174 | 0.01 |
| Peak | 0.3666 | 0.1192 | 0.002 |
| Number of nights with symptoms | 0.0250 | 0.0048 | <0.0001 |
Model R-Square = 0.1867, F-value = 17.18, Model p value <0.0001.
Comparison of Outcomes
The cost-consequence analysis in Table 6 indicates that the intervention group was more costly and no more effective than the control group. An asthma action plan alone may not be a cost-effective means to control the number of nights with symptoms and attacks in children with asthma. The median annual number of nights with symptoms was higher for the control group but the interquartile range was equal for both groups. The median and interquartile ranges for the number of attacks per year were equal for the asthma action plan and no asthma action plan groups. Overall, the health outcomes were not statistically significantly different between groups.
Table 6.
Cost-consequence analysis.
| Health outcomes | Action plan n = 217 | No action plan n = 618 | Increment | ||
|---|---|---|---|---|---|
| Adjusted total annual cost per patient (mean, 95% confidence interval) | $7,186 | $6,491–$7,882 | $6,326 | $5,999–6,652 | $860 |
| Number of nights with symptoms per year (median, interquartile range) | 0.00 | 1.71 | 0.60 | 1.71 | −0.60 |
| Number of attacks per year (median, interquartile range) | 4.00 | 6.00 | 4.00 | 6.00 | 0 |
All costs reported in 2003 Canadian dollars.
Sensitivity Analysis
Three one-way sensitivity analyses were performed on cost items with the most uncertainty, including inpatient care and productivity time loss (Figure 1). The incremental costs did not vary significantly from the base case for each sensitivity analysis. When the inpatient daily case cost was varied between the minimum and maximum, the incremental mean total cost per patient varied from CDN$811 to CDN$917 (95% confidence interval, CDN$449, CDN$1384). Adjustments for both sensitivity analyses in time loss values resulted in an incremental cost of CDN $853 (95% confidence interval, CDN$484, CDN$1221). In all sensitivity analyses, the total cost per patient remained greater in the intervention group compared with the control group.
Figure 1. One-way sensitivity analysis.
Vertical bar represents the base case incremental cost of $860. Points and box represent the incremental cost or range of costs under the assumption of the sensitivity analysis. Horizontal lines indicate 95% confidence intervals. All costs reported in 2003 Canadian dollars.
Discussion
The recommended management strategy for asthma consists of medications to control symptoms and underlying airway inflammation and a comprehensive education plan that promotes self-management through the control of environmental exposures to asthma triggers and through self-monitoring of symptoms (17). It has been recommended by the Canadian Consensus guidelines that a comprehensive asthma education program include an asthma action plan to help an individual monitor symptoms for adequate control (17). However, asthma action plans are often inadequately implemented by health practitioners owing to time constraints (14, 15).
The overall results for the cost-consequence analysis found no significant difference in effectiveness of an asthma action plan between the two groups. This finding is supported by a 2002 Cochrane Review on the effectiveness of written management instructions on asthma outcome measurements (25). This systematic review reported that none of the outcomes measured (e.g., hospitalization, ED visits, oral systemic corticosteroid use, lung function, days lost from school or work, unscheduled doctor visits, and respiratory tract infections) were statistically different when compared to no written plans for adults (25). However, a systematic review of several studies showed that asthma education strategies stressing self-efficacy strategies were effective in reducing the cost of health care utilization, improving physiological function, reducing morbidity, and increasing functional status compared to usual care in children (9).
Children with allergies have greater disease morbidity and a higher incidence of asthma attacks (26, 27). This was a consistent finding in the present study. The results showed that the intervention group reported a higher rate of allergies. After controlling for differences in disease severity, asthma control, season, and demographics, the annual total cost for the asthma action plan group remained greater than the control group by an average of CDN$860 per patient, primarily due to greater medication costs, more respiratory specialist visits, and higher inpatient care costs. It is possible that the intervention group was more adherent with medication regimens due to the use of the asthma action plan and hence, received better care and was more closely followed by a specialist compared to the control group.
Cost savings have been reported for educational intervention strategies that used peak-flow measurements for self-management compared with a symptom-based self-management plan in adult patients (7). A study evaluating the value of asthma action plans in children reported that 9 of 10 caretakers found the intervention to be useful in managing exacerbations (28). Moreover, Wolf et al. concluded that asthma self-management programs in children improved an array of outcome measures, but additional studies are necessary to compare morbidity and other quality indicators, such as functional status outcomes (9). Other research reported that self-management programs for children with severe asthma resulted in net savings, although their efficiency among children with mild or moderate disease was uncertain (16). Pediatric asthma self-management training and education were also medically effective and reduced costs associated with emergency room visits and hospitalizations (29). In contrast, the present study showed that the addition of an asthma action plan did not improve the rate of asthma exacerbations or frequency of nights with symptoms. Given that the cost of full asthma education was not substantial, the intervention would need to reduce other costs, such as inpatient care and parental productivity losses, to be cost-effective.
A number of limitations in the study must be considered. An important limitation was the cross-sectional nature of the data. This precluded inferences about causality, permitting only the examination of an association between the presence or absence of the intervention and costs and consequences (30). Another limitation was the risk of selection bias. Although a stratified sampling plan was used to recruit children with a range of asthma severity, a large proportion of the patients were recruited from emergency departments or specialty asthma clinics and thus the sample may represent a more severe subset of children compared to the general population of children with asthma. However, this is the subset of children for whom asthma action plans would potentially demonstrate the greatest benefit and are therefore important to study. Perhaps a greater limitation was the lack of a random allocation to the intervention or control groups. A greater proportion of the control group was selected from the ED compared to the intervention group. Thus, the control group may have been more representative of children who did not receive regular follow-up care, and therefore, would not have an opportunity to receive an action plan. It is also possible that the children who received asthma action plans did so because they have more severe asthma. The regression analysis of costs attempted to control for the major confounders of asthma severity and socioeconomic status, but this may have been imperfect. If selection bias related to asthma severity was not adequately controlled for and the intervention group had more severe asthma, then the study may have underestimated the incremental effectiveness of the asthma action plan in reducing asthma exacerbations and nights with symptoms and may have overestimated the incremental costs in this group compared with the control group. Another limitation was that the information was based on recall intervals for health care resources that ranged from 1 to 6 months and were linearly extrapolated to 1 year. There was a possibility that the information collected was subject to recall bias, although recall intervals with documented validity were selected (18). Finally, parents served as proxy reporters in this study. While parents may report use of their child’s health services reasonably well, it is expected that they may be less proficient at proxy reporting of symptoms and exacerbations. Guyatt et al. found only a weak correlation between a parent global symptom assessment and symptoms reported by children with asthma using a standardized quality-of-life instrument (31).
Conclusion
In this economic evaluation of asthma action plans, children in receipt of an action plan exhibited greater costs compared to a control group without demonstrating superior outcomes. Prospective, randomized controlled trials are necessary to assess the ability of written asthma action plans and personal management plans to mitigate deleterious outcomes in children with asthma and to determine if action plans should be targeted to a more severe subset of patients. In addition, research is required to establish the most cost-effective asthma education method for children, given the participation of parents as proxy reporters and educators.
Acknowledgments
Funded in part by AllerGen NCE Inc., a Network of Centres of Excellence Canada. The Networks of Centres of Excellence Canada is a joint initiative of the Natural Sciences and Engineering Research Council, the Canadian Institutes of Health Research, the Social Sciences and Humanities Research Council, and Industry Canada. W.J.U. is supported by a Canadian Institutes for Health Research New Investigator award. The authors’ work was independent of the funding source.
References
- 1.Millar WJ, Hill GB. Statistics Canada. Childhood Asthma Health Rep. 1998;10:9–21. [PubMed] [Google Scholar]
- 2.Doull IJ, Williams AA, Freezer NJ, Holgate ST. Descriptive study of cough, wheeze and school absence in childhood. Thorax. 1996;51:630–631. doi: 10.1136/thx.51.6.630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.GlaxoSmithKline Inc. Asthma in Canada: A Landmark Survey-Executive Summary. 2001. [Google Scholar]
- 4.Gibson PG, Ram FSF, Powell H. Asthma Education. Respir Med. 2003;97:1036–1044. doi: 10.1016/s0954-6111(03)00134-3. [DOI] [PubMed] [Google Scholar]
- 5.Ng TJ, Tan WC. Temporal trends and ethnic variations in asthma mortality in Singapore, 1976–1995. Thorax. 1999;54:990–994. doi: 10.1136/thx.54.11.990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang LY, Zhong Y, Wheeler L. Direct and indirect costs of asthma in school-age children. Prev Chronic Dis. 2005;2:A11. [PMC free article] [PubMed] [Google Scholar]
- 7.de Asis ML, Greene R. A cost-effectiveness analysis of a peak flow-based asthma education and self-management plan in a high-cost population. J Asthma. 2004;41:559–565. [PubMed] [Google Scholar]
- 8.Blaiss MS. Asthma disease management: a critical analysis. Ann Allergy Asthma Immunol. 2005;95(suppl 5):S10–S16. doi: 10.1016/s1081-1206(10)61004-6. [DOI] [PubMed] [Google Scholar]
- 9.Wolf FM, Guevara JP, Grum CM, Clark NM, Cate CJ. Educational interventions for asthma in children. Cochrane Database of Syst Rev. 2002;(4):CD000326. doi: 10.1002/14651858.CD000326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.The Nemours Foundation. [Accessed on 4 November 2006];KidsHealth for parents. Available from: http://www.kidshealth.org/parent/asthmabasics/dictionary/actionplan.html.
- 11.Lozano P, Connell FA, Koepsell TD. Use of health services by African-American children with asthma on Medicaid. JAMA. 1995;274:469–473. [PubMed] [Google Scholar]
- 12.Vollmer WM, Buist AS, Osborne ML. Twenty year trends in hospital discharges for asthma among members of a health maintenance organization. J Clin Epidemiol. 1992;45:999–1006. doi: 10.1016/0895-4356(92)90115-4. [DOI] [PubMed] [Google Scholar]
- 13.Wissow LS, Gittelsohn AM, Szklo M, Starfield B, Mussman M. Poverty, race, and hospitalization for childhood asthma. Am J Public Health. 1988;78:777–782. doi: 10.2105/ajph.78.7.777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Chapman KR, Ernst P, Grenville A, Dewland P, Zimmerman S. Control of asthma in Canada: failure to achieve guideline targets. Can Respir J. 2001;8(suppl A):S35–S40. doi: 10.1155/2001/245261. [DOI] [PubMed] [Google Scholar]
- 15.Jin R, Choi BC, Chan BT, McRae L, Li F, Cicutto L, Boulet LP, Mitchell I, Beveridge R, Leith E. Physician asthma management practices in Canada. Can Respir J. 2000;7:456–465. doi: 10.1155/2000/587151. [DOI] [PubMed] [Google Scholar]
- 16.Feenstra TL, Rutten-van Mölken M, Jager JC, van Essen-Zandvliet LEM. Cost-effectiveness of guideline advice for children with asthma: a literature review. Pediatr Pulmonol. 2002;34:442–454. doi: 10.1002/ppul.10177. [DOI] [PubMed] [Google Scholar]
- 17.Boulet LP, Becker A, Bérubé D, Beveridge R, Ernst P. Summary of the recommendations of the Canadian Consensus Conference on Asthma 1999. Can Med Assoc J. 1999;161(suppl 11):S1–S12. [PMC free article] [PubMed] [Google Scholar]
- 18.Ungar WJ, Coyte PC The Pharmacy Medication Monitoring Program Advisory Board. Health services utilization reporting in respiratory patients. J Clin Epidemiol. 1998;51:335–1342. doi: 10.1016/s0895-4356(98)00117-6. [DOI] [PubMed] [Google Scholar]
- 19.Ontario Ministry of Health and Long-Term Care. [Accessed on 4 November 2006];Ontario Case Costing Initiative [online] Available from: http://www.occp.com/
- 20.Statistics Canada. [Accessed on 4 November 2006];Wages, salaries and other earnings [online] Available from: http://cansim2.statcan.ca/cgi-win/cnsmcgi.pgm?Lang=E&ResultTemplate=Srch3&CORCmd=GetTCount&CORId=238.
- 21.World Health Organization. International Statistical Classification of Diseases and Related Health Problems, Tenth Revision–ICD-10. 2 [Google Scholar]
- 22.Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. N Engl J Med. 1992;326:862–866. doi: 10.1056/NEJM199203263261304. [DOI] [PubMed] [Google Scholar]
- 23.Krahn MD, Berka C, Langlois P, Detsky AS. Direct and indirect costs of asthma in Canada, 1990. Can Med Assoc J. 1996;154:821–831. [PMC free article] [PubMed] [Google Scholar]
- 24.Duan N. Smearing estimate: a nonparametric retransformation method. J Am Stat Assoc. 1983;78:605–610. [Google Scholar]
- 25.Written individualised management plans for asthma in children and adults. Cochrane Database Syst Rev. 2002;(3):CD002171. doi: 10.1002/14651858.CD002171. [DOI] [PubMed] [Google Scholar]
- 26.Guidelines for the Diagnosis and Management of Asthma - Update on Selected Topics 2002. NAEPP Expert Panel Report. 2002:132. doi: 10.1111/j.1745-7599.2003.tb00256.x. [DOI] [PubMed] [Google Scholar]
- 27.Guilbert TW, Morgan WJ, Zeiger RS, Bacharier LB, Boehmer SJ, Krawiec M, Larsen G, Lemanske RF, Jr, Liu A, Mauger DT, Sorkness C, Szefler SJ, Strunk RC, Taussig LM, Martinez FD. Atopic characteristics of children with recurrent wheezing at high risk for the development of childhood asthma. J Allergy and Clin Immunol. 2004;114:1282–1287. doi: 10.1016/j.jaci.2004.09.020. [DOI] [PubMed] [Google Scholar]
- 28.Dinakar C, Osdol TJ, Wible K. How frequent are asthma exacerbations in a pediatric primary care setting and do written asthma action plans help in their management? J Asthma. 2004;41:807–812. doi: 10.1081/jas-200038418. [DOI] [PubMed] [Google Scholar]
- 29.Halpin HA, McMenamin SB, Pourat N, Yelin E. An analysis of California Assembly Bill 2185: Mandating coverage of pediatric asthma self-management training and education. Health Serv Rep. 2006;41(Pt 2):S1061–S1080. doi: 10.1111/j.1475-6773.2006.00520.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Young K. Population health: concepts and methods. 2. Oxford: Oxford University Press; 2005. [Google Scholar]
- 31.Guyatt GH, Juniper EF, Griffith LE, Feeny DH, Ferrie PJ. Children and adult perceptions of childhood asthma. Pediatrics. 1997;99:165–168. doi: 10.1542/peds.99.2.165. [DOI] [PubMed] [Google Scholar]