The cost effectiveness of palivizumab in congenital heart disease: a review of the current evidence

Abstract

Objective:

To review and summarize the literature concerning the cost-effectiveness of palivizumab compared to no prophylaxis in infants and young children with congenital heart disease (CHD).

Methods:

A systematic literature search (MEDLINE to March 2012, limited to English language) identified studies that examined the cost-effectiveness of palivizumab in CHD populations. The quality of each study was assigned a quality score of 1–100 based on the Quality of Health Economic Studies (QHES) instrument.

Results:

Ten studies were identified through the search strategy, of which four principally addressed the research question and six additional articles examined CHD in conjunction with other high-risk indications for palivizumab in their economic analyses. QHES for the studies ranged from 58–100, with a median score of 93 (76 for principal articles, 94 for secondary analyses). Cost-utility analyses, which evaluated costs per quality-adjusted life year (QALY), showed favorable results in five analyses (range $10,329–$16,648 per QALY), while the other two suggested no cost-effectiveness ($146,061 and $169,971 per QALY). Of three cost-effectiveness analyses, which assessed costs per hospital admission prevented (HAP), two concluded that the drug was not cost-effective ($16,216/day of hospitalization prevented and $868,296/HAP), while one did not interpret the final result ($43,561/HAP).

Limitations:

Significant variance exists across study characteristics, analytic models utilized, duration of RSV seasons assessed, primary outcome measures evaluated, sensitivity analyses conducted, and other model assumptions. Further, it was difficult to obtain true CHD-based quality scores for the studies that analyzed more than one indication.

Conclusions:

The findings of this review currently remain inconclusive. Although a favorable trend was identified in the cost-utility analyses, additional rigorously conducted studies are necessary to better estimate the cost-effectiveness of palivizumab for CHD infants in clinical practice.

Keywords::

• Congenital heart disease (CHD)
• Respiratory syncytial virus (RSV)
• Palivizumab
• Cost-effectiveness

Introduction

Respiratory syncytial virus (RSV) is the primary cause of hospitalization for lower respiratory tract infection in infants and young children1. While virtually all children under 2 years of age become infected with RSV2, only some require hospitalization or medical care. Pre-term infants <35 weeks gestational age (GA) and those with chronic lung disease (CLD) and congenital heart disease (CHD) are at highest risk for severe RSV infection. Hospitalization rates have been reported as being up to 4–5-times higher in these groups compared to healthy infants3–7. Overall, CHD is the most prevalent underlying condition for bronchiolitis-associated hospital discharges among children under 5 years of age3.

Kristensen et al.4 conducted a population-based, matched case control study in Denmark from 1996–2003, to specifically identify risk factors for RSV hospitalization in infants with cardiac disease. Children <23 months had a RSV hospitalization rate of 5.65 per 100 child-years4. Down syndrome, cardiomyopathy, and, to a lesser extent, hemodynamically significant heart disease were important predictors of RSV hospitalization. Moreover, young age and cardiac decompensation predicted a more severe course of RSV infection.

CHD patients infected with RSV sustain more complications during their hospital course and may require both intensive care and mechanical ventilation8. Moreover, infants harboring an RSV illness who undergo interventions such as catheterization and cardiac surgery also develop increased medical problems perioperatively5,9 compared to those without CHD6,10,11. MacDonald et al.6 documented an RSV-related mortality of 37% in CHD compared to non-CHD patients (1.5%). Across 12 studies from 1966–2009, Welliver et al.12 reported an overall 2–37% mortality rate for CHD children infected with RSV, while the estimated case-fatality rate for infants without comorbidities or risk factors is less than 1%.

It should be noted that CHD infants, though classified as an independent risk group, are not homogenous. CHD patients can be further stratified according to type and severity of cardiac disease including diagnosis (i.e., cardiac defect) and patient care (i.e., uncorrected, surgery planned, partially corrected, etc.)13. For instance, among CHD patients, it has been reported that those with pulmonary hypertension, those requiring medication for congestive heart failure, and those with cyanosis were at the greatest risk for RSV infection14,15. Variations in structural cardiac disease can also directly impact risk for hospitalization, and therefore accurate documentation of specific disorders and severity of underlying illnesses are highly relevant and may influence the findings in individual studies.

In 1998, palivizumab (Synagis; MedImmune, Gaithersburg, MD, and Abbott Laboratories, Abbott Park, IL), a human monoclonal antibody, was approved for RSV prophylaxis by the Federal Drug Agency (FDA) in the US and subsequently in other countries world-wide. The approval was granted on the findings from the initial IMpact, randomized, placebo-controlled study16, which demonstrated a 55% relative risk reduction (RRR) in RSV-related hospitalizations in premature infants and those with bronchopulmonary dysplasia (BPD). Later, the safety, tolerability, and efficacy of palivizumab for children with hemodynamically significant CHD aged <2 years was also established in a multi-center, randomized, double-blind, placebo controlled trial in which a 45% RRR for RSV-related hospitalizations was reported14. Both the cyanotic and acyanotic sub-groups realized reductions in hospitalizations, although the power of the study was inadequate to determine differences among the sub-groups. Based on these results, in 2003 the American Academy of Pediatrics (AAP) revised their guidelines for RSV prophylaxis to include children <24 months with hemodynamically significant CHD17. In a non-inferiority cardiac trial7 which compared motavizumab, a second generation monoclonal antibody, to palivizumab, RSV-related hospitalizations were similar, but the former has not been approved by the FDA. To date, palivizumab remains the only approved prophylactic strategy to protect against RSV in high-risk infant groups, including those with CHD.

Despite the fact that prophylaxis with palivizumab has been shown to significantly reduce rates of RSV-related hospitalization, high product costs continue to stall widespread adoption. In 2010 CAD, costs were cited as $752.26 per 50 mg vial and $1504.51 per 100 ml vial18, and on average, infants receive five intramuscular injections throughout a single RSV season. Attempts have been made to quantify when and for whom palivizumab is cost-effective, but the results have varied as some studies have supported cost-effectiveness, while others have not or have remained inconclusive. Although analyses of the cost effectiveness of palivizumab specific to infants with cardiac disease are limited, some studies have indeed focused solely on this group. Thus, our primary objective was to review the existing literature of the economic evaluation of palivizumab for infants with CHD, given that it is a uniquely high risk group with substantial RSV-related morbidity and mortality that merits consideration for prophylaxis.

Patients and methods

A systematic literature search was conducted to identify studies (limited to English language) that examined the cost-effectiveness of palivizumab in CHD populations. A MEDLINE (Ovid) search was performed which included articles from 1950 to March 2012 using the following search terms: (‘respiratory syncytial virus’ OR ‘RSV’) AND (‘congenital heart disease’ OR ‘chd’) AND (palivizumab OR monoclonal antibody$) AND (‘costs and cost analysis’ OR econom$ OR cost effective$ OR cost$ OR pharmacoeconomic$ OR budget$). The search also included the Cost-Effectiveness Analysis registry and the UK National Health Service (NHS) Economic Evaluation Database (NHS EED). Studies that were included in this analysis evaluated the cost-effectiveness of palivizumab for cardiac disease patients, either independently or in conjunction with other high-risk indications (commonly prematurity and BPD). For the latter, the findings pertaining to CHD infants and children were extracted from these articles and included, if relevant. Cost-effectiveness, cost-utility, and cost-benefit analyses were considered for inclusion only when palivizumab was compared to no prophylaxis. All pharmacoeconomic evaluations had to include an Incremental Cost-Effectiveness Ratio (ICER). Studies that did not calculate an ICER, such as population-based studies that reported on hospitalization rates and associated costs, were excluded. Cost analyses and cost comparisons were also excluded. The results of the search strategy and articles retrieved are shown in Figure 1.

Figure 1.  Search strategy for identifying relevant literature.

Four reports met inclusion criteria for an economic evaluation of palivizumab compared to no prophylaxis solely in a cardiac disease population and six additional articles were identified that examined CHD in conjunction with other high-risk indications in their economic analyses. Each study was assigned a quality score (QS) of 1–100 based on the Quality of Health Economic Studies (QHES) Instrument19. The QHES is a valid tool designed specifically for the assessment of health economic studies, and criteria included in this instrument are considered to reflect overall quality of the analyses19. Base-case results were converted to 2012 USD to facilitate comparison using the Bank of Canada’s online currency converter20 and the US Department of Labor online inflation calculator21 in Table 1.

Table 1.  Study characteristics of cost-effectiveness analyses of palivizumab in CHD populations.

Reference	Population	Country	Perspective	Time horizon	Quality score
Studies that focused on CHD infants and children only
Yount and Mahle23	Infants with CHD	USA	Societal	Lifetime	84
Rackham et al.25	Infants <1 year with CHD	UK	Payer	Not specified	68
Nuijten et al.22	Infants <2 years with CHD	Germany	Societal (payer secondary)	Lifetime	94
Harris et al.24	Children <2 years with CHD	Canada	Societal	Not specified	58
Studies that included CHD in conjunction with other prophylaxis indications in their analyses
Nuijten et al.26	Pre-term infants ≤35 weeks GA, infants with CHD or BPD	UK	Payer (societal secondary)	Lifetime	94
Resch et al.27	Pre-term infants ≤35 weeks GA, infants with CHD or BPD	Austria	Payer (societal secondary)	Lifetime	91
Wang et al.30	Pre-term infants ≤35 weeks GA with and without CLD; infants with CHD	UK	Payer (societal secondary)	Lifetime	94
Nuijten et al.29	Pre-term infants 32–35 weeks GA, infants with CHD or BPD	Netherlands	Payer (societal secondary)	Lifetime	94
Hampp et al.31	Infants with various combinations of: CLD, CHD, prematurity (≤32 weeks GA), and those with none of these indications	USA	Payer	1 year	87
Resch et al.28	Pre-term infants ≤36 weeks GA, infants with BPD or CHD	Austria	Payer (societal secondary)	Lifetime	100

CHD, congenital heart disease; GA, gestational age; BPD, bronchopulmonary dysplasia; CLD, chronic lung disease.

Results

Characteristics of each study are reported in Table 1. Scores pertaining to quality ranged from 58–100, with a median score of 93. The median score of the four key articles was 76 compared to 94 for the six additional articles for which CHD was a secondary analysis. For those six studies that reported on CHD in conjunction with other high-risk infant groups, the quality scores reported in Table 1 pertain to the entirety of the article rather than the sub-section on CHD. Of the four reports that analyzed only CHD infants, one study22 concluded that palivizumab was cost-effective (QS 94), while two did not (QS 84, 58)23,24, and one25 did not specify a conclusion (QS 68). Of the remaining articles that included CHD as a sub-analysis in addition to multiple RSV prophylaxis indications, four reported favorable results of cost-effectiveness (QS 91–100, median 94)26–29, while two found that costs outweighed the benefits of prophylaxis (QS 87, 94)30,31 for this group. Results for both identified studies and subsidiary studies are presented in Table 2.

Table 2.  Results of cost-effectiveness of palivizumab in CHD populations.

Reference	Currency and year	Hospitalization rate		Discount rate	Base case ICERs	Base-case ICERs (2012 USD)	Sensitivity analysis range	Cost-effectiveness threshold	Cost effective?
		Palivizumab	No prophylaxis
Key studies that focused on CHD patients only
Yount and Mahle23	2002 USD	5.3	9.7	3%	$114,337/QALY	$146,061/QALY	$60,233–382,675/QALY	Not specified	No
Rackham et al.25	2002 USD	7	13	Not specified	$34,100/HAP	$43,561/HAP	None	Not specified	Not specified
Nuijten et al.22	€2006	5.3	9.7	5%	€9259/QALY	$13,532/QALY	€933–4098/QALY	€30,000/QALY	Yes
Harris et al.24	2007 CAN$	1.7	2.9	Not specified	$15,514/1 day of hospitalization prevented	$16,216/1 day of hospitalization prevented	$5281–20,421/1 day of hospitalization prevented	Not specified	No
Subsidiary studies that included CHD in conjunction with other prophylaxis indications in their analyses
Nuijten et al.26	£2003	5.3	9.7	3.5%	£6664/QALY	$13,623/QALY	£5508–58,685/QALY^a	£25,000	Yes
Resch et al.27	€2006	5.3	9.7	5%	€11,390/QALY	$16,648/QALY	€7694–20,624/QALY	€30,000	Yes
Wang et al.30	£2006	5.32	9.72	3.5 %	£79,800/QALY	$169,971/QALY	£91,000–123,300/QALY^b	£30,000/QALY	No
Nuijten et al.29	€2006	5.3	9.7	Economic: 4.0%; clinical: 1.5%	€7067/QALY	$10,329/QALY	€4215–12,972/QALY	Not specified	Yes
Hampp et al.31	2010 USD	1.98	3.63	Not Specified	$823,868/HAP	$868,296/HAP	$63,870–1,277,397/HAP^c	Not specified	No
Resch et al.28	€2010	5.3	9.5	5%	€8956/LYG €8484/QALY	$11,577/LYG $10,967/QALY	€5656–15,352/QALY	Not specified	Yes

ICER, incremental cost effectiveness ratio; QALY, Quality Adjusted Life Year; HAP, hospital admission prevented; LYG, life year gained.

^aSensitivity analysis was reported for pre-term/BPD patients; ^bcohort-based incidence rates (0.5% and 2.5%) are used in the sensitivity analysis; ^csensitivity analysis based on premature infants <6 months.

Identified studies

Studies that focused primarily on infants and children with CHD

Yount and Mahle23 examined the cost utility of palivizumab prophylaxis in children with CHD in the US using a decision-analytic model. The authors used a distribution of various heart lesions from a population-based study32 in creating their hypothetical patient cohort. Specific types of cardiac disease were taken into account in the inclusion criteria such as those who would likely meet criteria for diagnoses of cyanosis, congestive heart failure, and pulmonary hypertension. Rates of hospitalization, admission to the Intensive Care Unit (ICU), and length of hospital stay (LOS) were extrapolated from the Feltes et al.14 cardiac trial. All monetary units were reported in 2002 USD. Direct costs per infant using a 5-dose schedule of 100 mg vials were totaled at $6160 per RSV season. Authors estimated administration costs ($94/infant per RSV season) based on National Occupational Employment and Wage Estimates from the US Department of Labor. Hospitalization costs were based on a retrospective analysis of infants admitted to pediatric hospitals with RSV-related primary diagnoses. The average LOS was 4.5 days and average hospital costs were cited at $6789 or $1508/day. Costs of admission to the ICU were calculated at 2.5-times this cost. Indirect costs were also considered. The cost of missed work for one parent accompanying their child to the hospital was estimated at $1039 for a parent of an infant who received prophylaxis, compared to $1292 for a parent of a child who did not, based on the average LOS visits cited above. Further, in their model, a mortality rate of 3% for all infants admitted to the hospital was assumed. To measure cost-utility, authors used data consistent with a previously published study33 concerning adults with congestive heart failure (CHF) and assigned such utilities to CHD infants that result in long-term exercise limitations.

Results demonstrated unfavorable cost-effectiveness; prophylaxis for the palivizumab group resulted in a net loss of $20,415,753 for a hypothetical cohort of 10,000 CHD patients, when considering all costs. Moreover, the cost of administering prophylaxis to prevent 1 day of hospitalization was $4600, almost triple the cost of a single day spent in the hospital. The discounted ICER found was $114,337/Quality Adjusted Life Year (QALY) ($146,061/QALY 2012 USD) and sensitivity analyses ranged from $60,233–$382,675/QALY. Varying mortality rates had a large effect on cost per QALY saved. Overall, it was concluded that the cost of prophylaxis was high compared to the benefits.

In a retrospective cohort study, Rackham et al.25 identified patients with CHD and confirmed cases of RSV across three flu seasons from a hospital database in England. Patients included were <1 year of age, reflecting the recommendations of the British Pediatric Cardiac Association34. CHD patients were defined as those with hemodynamically significant lesions that would require surgery. The authors found a hospitalization rate of 10% for CHD infants admitted for RSV-related infection, comparable to that of the placebo group in the Feltes et al.14 randomized controlled trial (RCT). Applying the 45% RRR resulted in hospitalization rates of seven patients per year for the palivizumab group, and 13 patients per year for the placebo group. Costs per day for an admission to the cardiology ward and to the pediatric ICU were acquired from the Department of Trust at the hospital. All monetary units were reported in 2002 USD. The dosage and cost of palivizumab, obtained from the British National Formulary35, was based on the cost of a 100 mg vial of palivizumab in the UK in 2004, which was $1360. To treat the cohort of 131 identified patients at the single hospital with palivizumab (five doses per season) would cost an estimated $669,000 each year.

Palivizumab use contributed to reduction in costs by decreasing hospitalization days, including the ICU and cardiology wards, resulting in a savings of $368,200 each year. Other potential benefits to the patients and to the hospital were acknowledged, such as a decreased risk of nosocomial infection with fewer children admitted with RSV. Further, reduced hospital bed occupancy would contribute to cutting costs associated with inter-hospital transfer services, which were calculated based on ambulance transfer costs. Overall, the additional cost of providing palivizumab for these CHD patients was $204,500 per year. This amounted to $1540/patient, or $34,100/hospital admission prevented (HAP) ($43,561/HAP 2012 USD). The authors did not conclude whether or not this represented cost-effectiveness.

Nuijten et al.22 examined German infants <2 years with hemodynamically significant CHD, consistent with national recommendations for palivizumab prophylaxis. The authors used a decision-tree model in their economic analysis from the perspective of the German healthcare setting. Data from the Feltes et al.14 RCT were used to calculate both life years gained (LYGs) and QALYs attributable to palivizumab use. A lifetime horizon was used to provide an accurate assessment of the long-term effects of morbidity beyond the hospitalization period. Sources of input were clearly defined in calculating mortality rates (based on a CHD population of a meta-analysis conducted in the UK26), life expectancy, as well as utility scores. Use of palivizumab was based on an average of 4.93 doses per season. Inpatient costs were based on the total number of RSV hospitalizations according to the German DRG system, a national pricing system for hospitals. All monetary units were reported in €2006. Average cost of RSV-related inpatient treatments was approximated at €17,687 per hospitalization. Indirect costs, defined as productivity losses, were also calculated; work-loss costs for one parent accompanying a child to each visit amounted to €144 per season. Parent work-loss costs associated with child hospitalization were approximated at €1093 per hospitalization. The loss of productivity following RSV-related child mortality was included in long-term indirect costs and was based on lifetime productivity estimates for the average German citizen.

Palivizumab use in CHD infants resulted in an additional cost of €6972 per patient compared to no prophylaxis, considering both the cost of the drug and hospitalization. Scenario analyses factoring in the effects of asthma resulted in a reduction in the incremental cost. After discounting, the base case analysis resulted in ICERs of €10,116/LYG and €9529/QALY ($14,785/LYG and $13,532/QALY 2012 USD). The ICERs were less favorable when considering the perspective of the health insurance payer, based solely on direct costs. A scenario analysis that excluded mortality from the model largely improved the ICER. The authors concluded that, from both perspectives, palivizumab was cost-effective, as in most cases the ICER was below the accepted threshold of €30,000/QALY.

Harris et al.24 used a decision-analysis model based on a societal perspective to evaluate the cost-effectiveness of palivizumab for infants <2 years with hemodynamically significant CHD in British Columbia, Canada, that was consistent with recommendations from the Canadian Pediatric Society36 for prophylaxis. The authors retrospectively compared cases to non-prophylaxed historical controls. At the study site (pediatric hospital), a palivizumab prophylaxis program was already established and, thus, costs and outcomes were compared before (hypothetical control cohort) and after (palivizumab cohort) the institution of the program. For the palivizumab cohort, those who were ≥36 weeks GA at birth, <2 years of age at the onset of the RSV season, and had received the appropriate monthly injections of palivizumab were included. All monetary units were reported in 2007 CAN$. Direct costs of palivizumab were calculated based on the wholesale price in British Columbia at the time, which was $1467.81/100 mg vial. Mean number of doses administered during the RSV season was 4.5, and the total cost of prophylaxis per child per RSV season was $7915.62. Costs associated with administration were considered including hourly wages of medical staff. Hospitalization costs were estimated and obtained from the Provincial Health Services Association and totaled at $311,227 for the historical cohort vs $33,800 for the palivizumab cohort. Indirect costs were also calculated. Travel costs combined with the cost of a parent accompanying their child for prophylaxis amounted to $472.37 per family per RSV season. Work loss costs for a single parent accompanying their child to the hospital were estimated at $33,833.20 for the historical cohort and $4008.11 for the palivizumab cohort.

Hospitalization rates were 2.9% before the palivizumab prophylaxis program, compared to 1.7% following the program’s institution, resulting in a 42% reduction. In addition, the risk of admission to the ICU was reduced by 86% and total number of days spent in the hospital had an 83% reduction in the palivizumab cohort. The authors also reported that young age (<1 year old), presence of chromosomal abnormalities, and presence of feeding difficulties were all associated with an increased risk of hospitalization.

After accounting for all direct and indirect costs, the study authors reported an incremental cost to treat one child of $8292/RSV season, with 87% of this attributed to the cost of the prophylaxis itself. Considering all costs during the study period, the program resulted in a net loss of $2.4 million compared to no prophylaxis. The cost to prevent 1 day of hospitalization was $15,514 ($16,216 2012 USD), which was over 10-times the total cost of 1 day spent in the hospital. Sensitivity analyses showed that reducing the cost of the drug had large-scale impacts on cost-effectiveness. Based on their findings, the authors concluded that palivizumab was not cost-effective, though no official threshold has been established in Canada.

Studies that included CHD in conjunction with other prophylaxis indications in their analyses

Nuijten et al.26 conducted an analysis using a decision-tree model from the perspective of the UK National Health Service and also considered a secondary societal perspective. Hospitalization rates were derived from the overall CHD population in the Feltes et al.14 trial, with sub-stratifications of cyanotic vs acyanotic infants considered in some cases. Cost estimates associated with administration were based on 4.87 monthly doses of palivizumab/RSV season performed during a routine pediatric visit. For LOS, UK-specific data was used in computing an average of 12.39 days (6.25 for the pediatric ward and 6.14 for the ICU). Study authors used CHD data in calculating mortality estimates (2.3% for the palivizumab group vs 4.0% for the no prophylaxis group) and life expectancy estimates were appropriately derived from a UK-based CHD study37. All monetary units were reported in £2003.

Prophylaxis with palivizumab resulted in a discounted ICER of £6664/QALY ($13,623/QALY 2012 USD) for the CHD group. The cost/LYG of £7002 ($14,315/LYG 2012 USD) after discounting was slightly higher. Separate discounted ICERs were reported for CHD sub-populations; £14,816/QALY ($30,289/QALY 2012 USD) for the cyanotic and £3512/QALY ($7179/QALY 2012 USD) for the acyanotic sub-groups. A cost-effectiveness acceptability curve for palivizumab prophylaxis in all CHD patients showed a 0.98 probability that the ICER would be less than £25,000/QALY, an established threshold in the UK according to the study authors. The exclusion of asthma costs led to only a moderately higher ICER, and the inclusion of indirect costs led to a ‘dominant’ outcome in CHD patients.

In their hypothetical cohort of high-risk infants and children in Austria, Resch et al.27 assumed a lifetime follow-up period to account for the long-term effects of RSV-related morbidity and mortality. In their decision-tree model, base-case input data for the CHD group used hospitalization rates from the Feltes et al.14 trial. CHD mortality rates used in the analysis were pooled from the results of the placebo arms of three trials16,38,39. In calculating utilities, the authors extrapolated UK-based utility data to Austria. A slight reduction in life expectancy was applied to the CHD group, according to the results of a UK study37. For resource utilization and costs, the dose schedule for the CHD infant was assumed at 4.93 doses per season, consistent with the RCT. Administration costs assumed that palivizumab would be administered by a pediatrician, and costs per out-patient visit were calculated on a weighted average of official tariffs from four Austrian regions. LOS was based on Austrian data, though not for CHD patients specifically. Scenario analyses were based on LOS, applying Austrian ‘per diems’ for the pediatric and ICUs. Indirect costs such as loss of productivity were also calculated according to the human capital approach, which estimated lifetime productivity costs. All monetary units were reported in €2006.

In the CHD indication, use of palivizumab led to an additional cost of €4349, which decreased to €3724 when the cost of asthma treatment was included. The discounted ICER, which excluded asthma costs, was €11,390/QALY ($16,648/QALY 2012 USD). Cost-effectiveness became more favorable when asthma costs were considered, reducing the ICER to €9754/QALY ($14,256/QALY 2012 USD). For the secondary societal perspective that included indirect costs, the discounted ICER for the CHD group was €917/QALY ($1341/QALY 2012 USD). The results, particularly from the societal perspective, were the most favorable for CHD patients compared to other indications.

Wang et al.30 systematically reviewed the evidence of cost-effectiveness of palivizumab in addition to conducting their own analysis, using best-case estimates for the UK. The authors developed an individualized base-case decision-tree model, titled the Birmingham Economic Evaluation. Analyses were conducted both from the perspectives of the National Health Service (NHS) and society. For the CHD group, the Feltes et al.14 trial data were used for hospitalization rates in the analysis, but alternative data were used in the sensitivity analysis; those from a Swiss CHD cohort study40 which reported rates of 0.5% and 2.5%. Results from study authors’ previous meta-analysis of CHD-related studies found a mortality rate of 3.72%. It was assumed that palivizumab was routinely given in an outpatient setting, and appropriate times and wages were estimated to calculate administration costs. All monetary units were reported in £2006. Based on a five dose/season schedule, the direct cost of palivizumab was reported as £3736.90. With respect to hospitalization costs, the average length of ICU stay was adjusted for the CHD group at 6.14 days and non-ICU hospital stay was 6.25 days based on previous clinical trial data. Parent work-loss costs were considered, both in terms of loss associated with the administration of palivizumab (estimated at £209.70), and in terms of their child’s hospitalization (approximated at £1385.70). Authors acknowledged that utility data specific to CHD patients were lacking and, thus, extrapolated these values from the pre-term group.

From the NHS perspective, ICERs found were £79,800/QALY ($169,971/QALY 2012 USD), £78,400/LYG ($166,989/LYG 2012 USD), and £78,600/HAP ($167,415/HAP 2012 USD) for children with CHD. From the societal perspective, an ICER of £83,200/QALY ($177,213/QALY 2012 USD) was reported. In a separate analysis, which considered long-term sequelae such as asthma and recurrent wheezing, ICERs for the CHD patients were £64,600/QALY ($137,595/QALY 2012 USD) and £67,900/QALY ($144,624/QALY 2012 USD), from the healthcare and societal perspectives, respectively. While the ICER decreased in the case of considering long-term sequelae, overall, palivizumab was not found to be a cost-effective strategy in the CHD group. The lowest ICER found still exceeded a willingness-to-pay threshold of £30,000/QALY, established in the UK.

Nuijten et al.29 used country-specific inputs that were derived from the official Dutch costing manual and other relevant national data sources in conducting their analysis. A separate decision-tree model was created for CHD infants, due to their known elevated risk level and, accordingly, a small reduction in life expectancy was made for this group only. All monetary units were reported in €2006. For the base-case analysis, costs were calculated for 100 mg vials of palivizumab at €986 based on 4.93 doses per season. In terms of estimating hospital costs, the authors assumed those for the BPD indication to CHD patients, though they acknowledge this may be an under-estimation of the actual costs associated with this risk group. Estimates for costs associated with asthma were well defined based on previous literature and a national healthcare report. Work loss costs for parents as a result of their child’s hospitalization were approximated at €306, which included travel costs and fees for family physician consults. Long-term indirect costs, including lost productivity attributable to child mortality, were also considered based on average Dutch wage costs and were estimated at €40,133 annually.

Use of palivizumab led to an additional cost of €5926 per child. The inclusion of indirect costs saved €2670 compared to no prophylaxis. Palivizumab usage resulted in a discounted ICER of €7067/QALY ($10,329/QALY 2012 USD) considering total direct costs, including those associated with asthma. A probabilistic sensitivity analysis indicated a 90% chance that the ICER associated with palivizumab use would fall below €10,000/QALY in the CHD population. It was concluded that palivizumab was cost-effective in this article.

Hampp et al.31 conducted a decision-tree analysis to evaluate the cost-effectiveness across various indications of high-risk infants in a single, 6-month RSV season based on a Florida Medicaid population. Cost of palivizumab was estimated according to specific age group, and based on the Medicaid payment amounts that corresponded with National Drug Codes. Indication-specific data was used whenever possible, such as applying the Feltes et al.14 data to the CHD group. To estimate rates of hospitalization, an adjusted incidence rate was created based on the known reduction in hospitalizations attributable to prophylaxis. Incidence rates of RSV infection for CHD patients were reported as 1.98% in the palivizumab group, vs 3.63% in the no prophylaxis group. All monetary units were reported in 2010 USD. For the CHD group, the mean cost cited for an RSV-related hospitalization was $10,236, and an ICER of $823,868/HAP ($868,296/HAP 2012 USD) was found. The number needed to treat for this group was 61, indicating that 61 children had to receive prophylaxis for 6 months to prevent one RSV-related hospitalization.

Using a decision-tree model, Resch et al.28 provided an updated analysis from their 2008 cost-effectiveness report27 on multiple high-risk indications, one of which was CHD. All monetary units were reported in €2010. Both LYG and QALY outcomes were measured. Palivizumab administration costs were based on a mean of 3.98 injections, amounting to €3336/RSV season. Country-specific data was used in calculating costs, including compensation for diagnosis-based treatment groups. Administration costs considered both inpatient and outpatient-related costs for both physicians and specialists. Total hospitalization costs were €2494 for a mean LOS of 4.8 days according to the Austrian LKF-system. Production loss costs due to mortality were approximated using the human capital approach consistent with Resch et al.’s previous work. Epidemiologic data for the past 16 RSV seasons were considered in the creation of this hypothetical cohort, which was well defined.

Results showed discounted ICERs for the CHD group of €8484/QALY ($10,967/QALY 2012 USD) as well as €8956/LYG ($11,577/LYG 2012 USD). The ICER became more favorable when considering long-term costs associated with recurrent wheezing (€8253/LYG [$10,668/LYG 2012 USD] and €7818/QALY [$10,106/QALY 2012 USD]) for infants with CHD. The results of a secondary, societal perspective demonstrated superior cost-effectiveness, resulting in ICERs of €3214/LYG ($4155/LYG 2012 USD) and €3045/QALY ($3735/QALY 2012 USD). An additional scenario analysis which considered recurrent wheezing led to a decreased ICER, which was reduced even further upon the inclusion of indirect costs.

Discussion

There was no apparent relation between conclusions of cost-effectiveness and quality scores of the articles. This could be partly attributable to the heterogeneity of the study designs included in this review. In general, key studies had lower quality than subsidiary studies, though quality of the CHD component was more difficult to assess in the latter. In many cases, points for quality were deducted for failing to report on the direction and magnitude of potential biases22–24,26,27,29,30. Some studies did not provide a statement disclosing sources of funding23,25,27. Other areas where analyses could be improved pertained to the choice and justification of the model used in the study23,24, the analytic timeline of the analysis (i.e., only accounting for one RSV season), and conducting discounting24,25,31, stating, or justifying primary outcomes measures24,25, conducting a sensitivity analysis and performing an ICER25, and providing best-case estimates24. Thus, although overall quality was high, some elements were infrequently reported.

In general, analyses reported non-uniform results pertaining to the cost-effectiveness of palivizumab for infants and young children with CHD. Critical factors that must be considered when interpreting results include but are not limited to: perspective of the analysis, type of economic model(s) used, outcomes measured, cost considerations and definitions, and whether or not funding sources were disclosed. Moreover, analyses should be nested within the contexts of national healthcare systems and country-specific cost-effectiveness thresholds, and must also envelop geographical and cultural factors such as level of accessibility to medical care.

Of the four key studies, both Harris et al.24 and Yount and Mahle23 found poor cost-effectiveness of palivizumab within North American contexts, albeit being conducted a decade apart and in separate national healthcare systems. Yount and Mahle recognized their bias towards cost-utility as they estimated children with CHD to have a regular life span expectancy. Further, extrapolating exercise limitation data from CHF adults to CHD infants does not provide for a ‘best-case’ comparison, though they acknowledged this as a limitation. Finally, in that analysis, the mortality rates from the Feltes et al.14 RCT were not used, and did not differ between treatment groups. The Canadian study accounted for RSV admissions at a single institution and, like Yount and Mahle, did not consider costs associated with outpatient RSV-related treatment. Harris et al. also failed to calculate an ICER/QALY for the treatment intervention, the preferred unit of measure stipulated by the Canadian Agency for Drugs and Technological Health, which restricts the comparison of their data to other published studies. Further, the authors only found one death in their historical cohort, which does not provide for a representative estimate of generalizable mortality across the CHD population.

In contrast, Nuijten and colleagues’26–29 analysis of German infants demonstrated similar results to other European studies that have deemed palivizumab to be cost-effective across multiple high-risk indications, including CHD infants. The authors indicated that palivizumab is assumed not to have an impact on the severity of RSV infection. Findings showed that the inclusion of long-term indirect costs, such as lost productivity following the death of a child, positively impacted cost-effectiveness.

The final key study, by Rackham et al.25, limited their infant population to <1 year, as opposed to <2 years in most other CHD studies. They did not specify a time horizon for their study, nor a cost-effectiveness threshold to support an ultimate conclusion of cost-effectiveness. Instead, individual hospital Trusts were called on to make such determinations.

Results of the studies that included CHD as a sub-group also demonstrated variance. Many of these high quality studies considered the effects of asthma as well as lifetime productivity, both relevant factors in the long-term context of RSV-related impact26–29. Thus, consistent with the findings of Nuijten et al.22, these studies considered lifetime productivity losses attributable to RSV-related mortality from a new, analytical perspective. Other studies, such as the one by Wang et al.30, did not consider the effects of asthma or wheezing in the base-case and, rather, considered these in a separate analysis, though based on CLD data. Moreover, Hampp et al.31 did not account for the effects of asthma or mortality at all. In that analysis, it was acknowledged that some RSV-related hospitalizations could potentially not have been coded, and thus under-represented in their data set. In addition, only direct costs from the perspective of Florida Medicaid were considered.

In most cases cost-utility analyses, which evaluated costs per QALY, showed favorable results compared to cost-effectiveness analyses, which assessed costs per LYG or HAP. Further, the perspective of the analysis (societal vs payer) had an impact on conclusions. Most studies used input data from the 2003 Feltes et al.14 RCT for their models. Thus, hospitalization rates of 5.3 (with palivizumab prophylaxis) and 9.7 (placebo), and mortality rates of 3.3 (palivizumab) vs 4.2 (placebo) were assumed in most analyses. As noted in the studies that did not use those data, even a slight variance in mortality rates can have large-scale impacts on cost-effectiveness results. While mortality rates vary across sub-groups of high-risk RSV infants, CHD patients have the highest risk. Further, the majority of studies corrected for a small reduction in life expectancy for children with CHD, based on the work of Wren and O’Sullivan37. That UK study found that 82% of children with CHD survived to 1 year, and 78% were predicted to live until the age of 1622. To assess utilities, most analyses referred to the Greenough et al.41 study, which reported utility scores for children with histories of RSV infection using the Health Utility Index (HUI). As described above, analyses that considered long-term sequelae such as asthma and wheezing demonstrated that these are important cost-drivers.

Conclusions

While the studies included in this review constitute an important first step, they are mostly based on pharmacoeconomic modeling, which provide representations of chronic disease and are limited by the many assumptions embedded within these models. The studies that were not based on modeling included a pre–post comparison24 and a retrospective cohort analysis25, yet both of these articles had limitations, as previously addressed. While models are timely and cheap, systematic validation of disease models in the real world and their impact on resource utilization are still necessary. Rigorous, prospective collection of data costs, resource utilization, and outcomes are needed to assess the true cost-effectiveness of palivizumab and the total budget of prophylaxis in this population.

Overall, quantifying the cost-effectiveness of palivizumab is challenging. Authors report non-uniform results, such that conclusions need to be interpreted within the independent contexts of each analysis. Definitions of what constitutes cost-effectiveness vary, as do accepted willingness-to-pay thresholds, which may not be established in certain countries. A threshold of $50,000/QALY is commonly accepted in North American settings, while that in the UK has been estimated as £25,000–30,000/QALY. Feltes and Sondheimer8 addressed the ongoing controversy surrounding the cost-effectiveness of palivizumab in high-risk infants, as well as the challenges in providing accurate measurements of both direct and indirect costs. The results of the current review reinforce those judgments.

It is well known that cardiac disease patients have elevated risks of morbidity and mortality with both community acquired and nosocomial RSV-related illness and, consequently, their families are faced with increased financial burdens related to hospitalization and surgery. Since few studies have rigorously examined the cost-effectiveness of palivizumab in infants with cardiac disease, it is evident that additional analyses are required to help inform, and potentially guide, recommendations for clinical practice. While alternatives to prophylaxis such as vaccines are being explored, palivizumab remains the only approved strategy to protect against RSV to date. Given its expense, careful scrutiny in selecting optimal patient groups for prophylaxis is imperative.

Transparency

Declaration of funding

This study was not funded.

Declaration of financial/other relationships

KL and BP have disclosed that they have received grants from and are on the Speakers’ Bureau for Abbott and MedImmune. Both KL and BP have previously acted as consultants to Abbott. KL and JH have disclosed that they are employees of Sunnybrook Health Sciences Centre. BP is an employee of McMaster University.