 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
From Children's Healthcare of Atlanta and Department of Pediatrics, Division of Cardiology, Emory University School of Medicine, Atlanta, Georgia
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods. Probabilities of hospitalization and efficacy of prophylaxis were based on published results. Costs of hospitalization were derived from a published analysis of bronchiolitis hospitalization costs from a consortium of children's hospitals. A hypothetical cohort of 10 000 CHD patients (half of whom would receive palivizumab) was created to calculate cost-savings and cost-utility. To assess cost utility, we assumed that by reducing hospitalization, palivizumab would reduce RSV-related hospital mortality, generally reported to be 3% in CHD patients. Sensitivity analysis was performed.
Results. On the basis of a protocol of 5 monthly doses of palivizumab, the cost of prophylaxis for 1 RSV season was calculated as $6160 per patient. After accounting for impact on direct and indirect costs of hospitalization, administration of palivizumab to 5000 CHD patients would result in a net loss of $20 415 753. If one assumes that palivizumab confers a survival benefit, then the cost of life-year saved is $100 338 and cost of quality-adjusted life-year saved is $114 337.
Conclusions. The cost of palivizumab prophylaxis was high relative to benefits realized. Given the large number of CHD patients who might be considered candidates for RSV prophylaxis (>6000 patients per year in United States) routine use of palivizumab in young children with hemodynamically significant CHD needs to be evaluated further.
Key Words: bronchiolitis • congenital heart disease/defects • cost analysis
Abbreviations: RSV, respiratory syncytial virus • CHD, congenital heart disease • AAP, American Academy of Pediatrics • ICU, intensive care unit • CHF, congestive heart failure • QALY, quality-adjusted life-year
Bronchiolitis caused by respiratory syncytial virus (RSV) is an important cause of morbidity and mortality in the pediatric population, accounting for 
100 000 hospitalizations per year in the United States.1 Children who are hospitalized for RSV without other risk factors experience an 1% mortality.1 Children with congenital heart disease (CHD) are more vulnerable, having a mortality of 2% to 3% for RSV-related hospitalizations in some studies.1,2
Palivizumab (Synagis; MedImmune Inc, Gaithersburg, MD) is a monoclonal antibody that has been shown to lessen the severity of RSV bronchiolitis.3 In a recent multicenter, placebo-controlled trial, palivizumab significantly reduced the number of RSV-related hospitalizations in children with CHD.4 This study demonstrated that the risk for hospital admission for RSV bronchiolitis was lower and the total number of hospital days was less in patients who received palivizumab. No survival benefit was demonstrated. On the basis in part of the results of this study, the American Academy of Pediatrics (AAP) recommended that "palivizumab is appropriate for infants and young children with hemodynamically significant CHD."5
The high cost of palivizumab remains a concern, however, and several studies have questioned the cost-effectiveness in preterm infants.6–8 The economic implications of palivizumab prophylaxis in the CHD population have not been evaluated. Given the high cost of RSV prophylaxis and large number of children who might be eligible on the basis of AAP recommendations (>6000 children in the United States) a critical examination of the costs of palivizumab in the CHD population is warranted.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Using the societal perspective, we considered medical costs and work-loss costs. The primary outcome was hospital savings. This represents the reduction in hospitalization cost minus the cost of palivizumab prophylaxis. Costs are reported in 2002 US dollars.
Data and Assumptions
The probability of hospitalization for patients who received palivizumab was 5.3% and for patients who did not receive prophylaxis was 9.7%.4 The number of days in intensive care was also higher for the placebo group: 71.2 versus 15.9/100 children.
Direct Costs
Medical Costs
The average wholesale acquisition cost of on a 100-mg vile of palivizumab was $1369.14 The recommended dose of palivizumab is 15 mg/kg administered once monthly throughout the RSV season. Assuming a mean weight of eligible children of 6.0 kg., the mean dose per patient would be 90 mg. We assumed that each infant received 5 doses and that no drug wastage occurred. Thus, the total direct per-infant cost of palivizumab per month was assumed to be $1232 (90 mg x $1369/100 mg) and for a single RSV season was $6160 ($1232 x 5 doses).
Administration Costs
Palivizumab is routinely administered in an outpatient setting. We assumed that the first visit would require 30 minutes of physician time to see the patient and discuss RSV prophylaxis along with 30 minutes of nursing time. The other 4 visits could be managed by nursing staff and would require 30 minutes per visit. Using 2002 National Occupational Employment and Wage Estimates from the US Department of Labor, we estimated pediatrician and nursing mean hourly wages at $68/hour and $24/hour, respectively.15 Costs for physician wages therefore would be $34 (1 visit x 0.5 hour/visit x $68/hour) and nursing costs of $60 (5 visits x 0.5 hour/visit x $24/hour). We considered overhead allocation and office staff time to be negligible, thus biasing toward prophylaxis. Total administration costs per infant per RSV season totaled $94.
Hospitalization for RSV
Cost estimates were obtained from a retrospective analysis of infants who were admitted to 10 children's hospitals with a principal diagnoses of bronchiolitis (International Classification of Disease, Ninth Revision code 466.1) and RSV pneumonia (International Classification of Disease, Ninth Revision code 480.1).16 The study included children with CHD (7.8%), preterm birth (19.6%), history of wheezing (9.3%), and history of hospitalization (11.3%). The 10 centers all were tertiary care hospitals with pediatric residency programs and pediatric ICUs. In this series, the average length of stay was 4.5 ± 4.2 days, and average hospital costs were $6789 or $1508/day ($6789/4.5 days). Admission to an ICU resulted in greater costs. In our present analysis, we adjusted number of ICU days with ICU costs being calculated at 2.5 x non-ICU costs on the basis of published pediatric and adult literature.17,18 For comparison, cost of RSV admission was determined from 2 additional sources: the Nationwide Inpatient Sample and a previous report that analyzed health maintenance organization data.7,19,20 The cost per day of RSV hospitalization from these 2 additional sources were $1675 and $1607, respectively, comparable to the 10-center data used for calculations in the present study.
Indirect Costs
Administration of Prophylaxis
For each office visit, we assumed that 1 parent accompanied the infant and that 3 hours of work would be missed. Using the US Bureau of Labor and Statistics 2002 average wage of $17/hour, lost wages during 1 RSV season would have an indirect cost of $255 (3 hours/visit x 5 visits x $17/hour). In addition, travel costs were estimated at $54 using the 2002 Internal Revenue Service mileage reimbursement rate of $0.365 per mile and assuming that on average a family traveled 30 miles round trip to each office visit ($0.365/mile x 30 miles x 5 visits).
Hospitalization for RSV
For each hospitalization, we assumed that 1 parent stayed with the infant at all times. We calculated the average length of stay for the treated group as 10.7 days and 13.3 days for the group that was not treated. Considering that 5 of every 7 days are workdays, the estimated work-loss cost therefore was $1039 for a parent of an infant who received prophylaxis ($17/hour x 8 hours/day x 10.7 days/hospital x 5/7) and $1292 ($17/hour x 8 hours/day x 13.3 days/hospital x 5/7) for a parent of a child who did not receive treatment.
Probability of Death, if Hospitalized for RSV
The multicenter trial reported by Feltes et al4 did not demonstrate a significant difference in mortality rate (3.3% vs 4.2%; P = .46) for patients who received palivizumab versus placebo. There also was not a significant difference in RSV-related adverse events. However, the aforementioned trial was designed to detect a difference in hospital admissions and may not have been sufficiently powered to detect a difference in RSV-related mortality. Previous studies have shown, however, that mortality for children who have CHD and are hospitalized for RSV bronchiolitis is higher than for children without CHD. Historical studies reported mortality of 37% for patients who had CHD and were admitted with bronchiolitis.21 Recent studies, however, have reported that hospital mortality for children who have CHD and are admitted with RSV bronchiolitis is between 1.6% and 3.4%.1,2 For purposes of cost-utility analysis, we constructed a model that assumed that the mortality rate for patients who are admitted to the hospital would be 3% whether they had received palivizumab or not. This assumption is based in part on the finding that the for patients who have CHD and were admitted for RSV bronchiolitis, there was no significant difference in need for ICU admission or need for mechanical ventilation between those who received palivizumab and control subjects (P = .98 and P = .88, respectively). Therefore, the difference in hospital admission rate for RSV bronchiolitis between the 2 groups (5.3% vs 9.7%) would result in a 0.13% reduction in overall mortality in our model.
Cost-Utility Analysis
Utility data (to evaluate the quality of life) in children and adults with CHD are lacking. However, to make comparisons between palivizumab prophylaxis and other public health interventions, we extrapolated data from adults with congestive heart failure (CHF) to the CHD population. Although the prevalence of CHF in CHD survivors is not well defined, there is abundant literature on exercise performance in adolescent and adult CHD patients. For the purposes of this study, we assigned an adult CHF utility to patients with CHD lesions that result in long-term exercise limitation as previously reported.22 The lesions with long-term exercise limitations are indicated in Table 1. Patients with lesions such as ventricular septal defect (with no significant long-term exercise limitations) are assigned a utility of 1.0. The utility for CHF was based on the Beaver Dam Health Outcomes Study, which evaluates health with the method of time tradeoffs.23 The utility of CHF from this study is 0.71. Because the utility of CHF can varying on the basis of method, we varied the CHF utility from 0.5 to 1.0 in the sensitivity analysis.
|
Discount Rate
All future costs and benefits are discounted at a rate of 3%. Discounting is used to calculate life-years lost.
Sensitivity Analysis
For cost-utility analysis, we evaluated the sensitivity of the model to variations in key assumptions over various ranges. We varied the hospital mortality for RSV bronchiolitis in CHD patients from 1% to 5%. We also varied the utility of CHF (exercise limitation) in patients with CHD from 0.50 to 1.0. The published data have not shown that palivizumab prophylaxis reduces hospital mortality for patients who have CHD and are admitted with RSV bronchiolitis.4 However, because the aforementioned study was not designed to detect such a benefit, we explored the possibility that palivizumab might reduce hospital mortality by 0% to 50% in the sensitivity analysis.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A number of previous investigators have attempted to analyze the cost-effectiveness of palivizumab in other high-risk groups. Greatest attention has been focused on very low birth weight infants, who have been reported to have a higher hospital mortality than CHD patients.26 Even in patients with relatively high hospital mortality and preexisting parenchymal lung disease, a number of investigators have questioned the cost-effectiveness of palivizumab.7,8,27,28 Joffe et al7 reported that the cost of life-year saved ranged from $110 000 to $1 200 000 (1995 US dollars). However, they reported that administration of palivizumab to a high-risk subset of patients (those who are born at <32 weeks’ gestation and require supplemental oxygen 
28 days) yielded a cost of $33 000 per life-year saved. Roeckl-Wiedmann et al8 analyzed the cost implications of RSV prophylaxis for premature infants. This study reported that the cost per hospitalization avoided were Euro 25 288 to Euro 204 684 (US currency: $25 529 to $198 543). Even in the highest risk groups, palivizumab did not result in hospital cost savings.
A number of criticisms have been put forth regarding the cost-effectiveness studies addressing palivizumab. Most studies, including our own, have not included the additional costs of RSV bronchiolitis that is treated on an outpatient basis. This is attributable in part to the difficulty in differentiating RSV upper respiratory infection from other viral entities. As such, data regarding the outpatient costs of RSV are limited. A recent analysis of RSV bronchiolitis in Canada suggested that expenditures for ambulatory patients accounted for 38% of direct costs.29 However, <1% of the entire infant population with an RSV infection required hospital admission. Therefore, on a per-patient basis, the direct costs of a single outpatient RSV infection is <2% of the cost of an RSV-related hospital admission and composes a relatively small portion of the overall direct costs in a targeted high-risk group such as the CHD population. The indirect costs of an outpatient case of RSV are less well defined and represent a limitation of the current study design.
An additional element of cost-effectiveness analysis engendering controversy has been the determination of indirect cost for hospitalization of a child. Nearly all analysis will include the impact of lost wages during hospitalization. However, there might also be additional lost wages after hospitalization. Leader et al30 suggested that the total average economic burden for a family per bronchiolitis admission was $4517 for premature and $2135 for term infants, including the value of lost productivity but excluding inpatient hospital and physician bills and lost income. An alternative approach is to perform contingent valuation of willingness-to-pay estimates of an individual, such as the value that the parent places on avoiding a hospitalization. In a previous study, Robbins et al31 reported that the parental willingness-to-pay estimate to avoid an RSV-related admission ranged from $1325 to $8700. However, concerns have been raised about the validity of such responses, especially by noneconomists.32
In the present study, we undertook a cost-utility analysis. The advantage of such an approach is that it allows one to make comparisons with other public health interventions. Unlike in many adult conditions for which there is abundant and reliable actuarial data, the life expectancy for children who were born with CHD in 2003 cannot be easily defined. In the present study, we assumed a normal life expectancy for all CHD patients. Although published data suggest that survival for some lesions such as repaired ventricular septal defect may be normal or near normal, other lesions have been associated with a reasonable risk for childhood mortality.1 Children with hypoplastic left heart syndrome and other forms of single ventricle have been reported to have a median survival of <10 years in some studies.33 In addition, utility data analyzing the quality of life in children with CHD are lacking. For the purposes of this study, we attempted to link long-term exercise intolerance to adult measures of utility in the setting of CHF.34 Although such assumptions have clear limitations, it should be recognized that in the present cost-utility analysis, the majority (58%) of patients were considered to have a utility of 1.0, and varying utility assumption did not profoundly alter cost of QALY saved results.
Other potential beneficial effects of palivizumab beyond its impact on length of hospital stay must be addressed in a cost-utility analysis. The present study assumes that palivizumab does confer any benefit outside the single RSV season. It is possible that patients with CHD might have adverse long-term effects of an RSV infection. It has previously been suggested that open-heart surgery be delayed 6 weeks after an RSV infection.35 Other investigators suggested that RSV bronchiolitis frequently resulted in the need to delay surgical procedures, which might result in poor growth and protracted cyanosis. It is interesting that in the multicenter trial, there was no association between palivizumab prophylaxis and the need to delay or cancel heart surgery. Whether an RSV infection has long-term implications for lung function in the CHD population is not known. It has been shown, however, that pulmonary function abnormalities are seldom determinants of later functional capacity in the CHD population.36
Although CHD is relatively rare (8/1000 live births) and the number of CHD patients who might be candidates is even less, the economic implications of the AAP Committee recommendations do have significant implications. Assuming that each year in the United States 3000 children who have critical CHD might be candidates for palivizumab and it may be administered into the second year of life, the overall cost of medication alone in the United States in 1 year could exceed $30 000 000. In the current economic climate, careful appraisal of the economic implications of palivizumab is mandatory. Additional data derived from the industry-sponsored trial, such as identification of potentially high-risk lesions such as single ventricle patients or defining a subset of heart patients such as those with acyanotic lesions, may help to refine the AAP recommendations and make better use of this potentially life-saving yet costly resource. In addition, given the relatively widespread use of palivizumab in the CHD population, it will be important to analyze how well practitioners are following published guidelines regarding RSV prophylaxis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Reprint requests to (W.T.M.) Children's Healthcare of Atlanta, Emory University School of Medicine, 1405 Clifton Rd NE, Atlanta, GA 30322-1062. E-mail: wmahle{at}emory.edu
No conflict of interest declared.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. de Lissovoy, L. S. Matza, H. Green, M. Werner, and T. Edgar Cost-effectiveness of Intrathecal Baclofen Therapy for the Treatment of Severe Spasticity Associated With Cerebral Palsy J Child Neurol, January 1, 2007; 22(1): 49 - 59. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
