Cost-effectiveness of Palivizumab for Respiratory Syncytial Virus: A Systematic Review

We conducted our systematic review by following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.¹⁸ The search strategy was developed with a Public Health Agency of Canada librarian. We conducted a scientific literature search for English- and French-language studies published in 3 electronic databases: Medline and Epub Ahead of Print and Medline In-Process & Other Non-Indexed Citations (Ovid interface), Embase (Ovid interface), and the Cochrane Library, which included the Health Technology Assessment Database, the National Health Service (NHS) Economic Evaluation Database, and the Database of Abstracts of Reviews of Effects. In our search, we used medical-subject headings and text words related to the following concepts: respiratory syncytial virus, palivizumab, economic evaluations, and cost-effectiveness. The primary search strategy was developed in Medline and adapted to other databases to account for database-specific vocabulary and functionality. A complete list of search terms and the full search strategy for Medline are summarized in Supplemental Table 4. We manually searched the reference lists from relevant articles and systematic reviews.

The protocol and eligibility criteria for studies are published on PROSPERO (identifier CRD42018104977). We included full economic evaluations (eg, cost-benefit analysis, cost-effectiveness analysis, and cost-utility analysis) in which palivizumab prophylaxis for RSV was compared with any comparator (eg, no prophylaxis) for infants up to 24 months of age on the basis of current guidelines from Canada’s National Advisory Committee on Immunization.¹⁹ We included economic evaluations that were conducted in OECD countries between 2000 and present and reported outcomes related to an incremental ratio of cost per unit (eg, cost per quality-adjusted life-year [QALY], cost per case averted, cost per life-year gained [LYG], and cost-benefit ratio). We excluded cost-minimization studies, cost-of-illness studies, and budget-impact analyses. We excluded studies conducted outside of OECD countries, studies published in a language other than English or French, and studies published before 2000.

Screening, data extraction, and quality appraisal were completed in duplicate (by S.M. and A.S.). All levels of screening were completed using DistillerSR (Evidence Partners, Ottawa, Canada). Conflicts were discussed and resolved through consensus. Data extraction was guided by Consolidated Health Economics Evaluation and Reporting Standards statement.²⁰ We collected study characteristics (publication year, country, study design, study perspective, time horizon, discounting, primary and secondary outcomes, use of cost-effectiveness thresholds, and funding sources), study population characteristics (age range, gestational age, health conditions, and setting), key parameters (RSV incidence and/or hospitalization rates, mortality rates, sequelae, cost of palivizumab, and number of doses), and results (base-case incremental cost-effectiveness ratios [ICERs], scenario analyses, type of sensitivity analysis, and influential parameters). The quality of included studies was assessed by using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Economic Evaluations.²¹ We classified a study as high quality if it met >80% of the JBI checklist criteria.²² 

We descriptively summarized the study characteristics and population characteristics. Cost-effectiveness outcomes were adjusted to 2017 US dollars (USDs) by using purchasing power parity rates from the OECD²³ and US inflation rates from the US Department of Labor. Unadjusted and adjusted ICERs were summarized. We conducted subgroup analyses to summarize the cost-effectiveness for studies conducted from remote regions of the Canadian Arctic and studies in which cost-effectiveness was reported in costs per QALY for preterm infants. For studies that included preterm infants, we stratified on the basis of gestational age at birth (weeks) and plotted this against the adjusted ICERs to visually identify the spread of ICER estimates and possible trends related to gestational age. The number of estimates and the proportion of them being cost-effective at various thresholds were summarized. A meta-analysis of cost-effectiveness was inappropriate because of the heterogeneity of the study setting, model designs, parameters used, population, and perspective taken in the studies.

Our systematic literature search identified 237 deduplicated records, of which 30 met our eligibility criteria and were included in our review (Fig 1).^14,24,–52 Conclusions of 2 studies^31,51 were updated by using more recent data,^32,39 which excluded them from our review’s analysis and conclusions. The 28 studies included were published between 2000 and 2018, with most conducted in the United States (n = 6), Canada (n = 5), Netherlands (n = 3), United Kingdom (n = 3), and Spain (n = 3). The rest of the studies were conducted in Austria (n = 2), Germany (n = 2), Italy (n = 1), Mexico (n = 1), New Zealand (n = 1), and Sweden (n = 1). Study characteristics are summarized in Table 1.

Most studies (83%) met >80% of the JBI quality appraisal checklist criteria (Supplemental Table 5). The 2 checklist items that were least met were whether the study results included all issues of concerns to users (39%) and whether all relevant costs and outcomes were identified (75%). Overall, studies included in this review were considered relatively high quality (Fig 2).

In 14 studies, the subject's chronological age was explicitly reported to be <24 months, whereas in the other 14 studies, it was assumed that the cost-effectiveness of palivizumab was assessed in infants <24 months of age on the basis of their respective country guidelines on palivizumab use. High-risk infant populations were often studied, and, in some cases, overlapped: preterm infants (≤35 weeks’ gestational age [wGA]) (n = 19), BPD or CLD (n = 13), CHD (n = 11), and other risk factors (n = 6).

Base-case analyses were almost equally conducted from a societal perspective (n = 13) or health care–payer perspective (n = 15). In 8 of the 15 payer-perspective studies, additional analyses were performed from a societal perspective. Time horizon ranged from 6 months to lifetime; a time horizon was not reported in 1 study.⁴⁸ (Table 1) Discount rates ranged between 3% and 5%; in 5 studies, the authors did not discount because of a limited time horizon,^{24,33,34,45,46} and a discount rate was not reported in 3 studies.^41,48,49 The majority of studies were industry sponsored (n = 17; 61%). Cost-effectiveness was mostly reported as cost per QALY (n = 20) and cost per hospitalization averted (HA) (n = 5). For the remainder of this review, we described results using adjusted ICERs (2017 USDs); original unadjusted ICERs are summarized in Table 2.

For studies in which cost-effectiveness was reported in cost per QALY units, we summarized the number of estimates, the ICER ranges, and the proportion of estimates under selected thresholds of $50 000 to $200 000 per QALY, stratified by population subgroups and study perspective in Table 3. From a health care–payer perspective, there were 22 varying cost-effectiveness estimates for preterm infants, ranging between $5188 and $791 265 per QALY.^* The subgroups with the next highest estimates were preterm infants with risk factors (n = 14),^25,37,39,40 in which the ICER was between $177 and $169 103 per QALY; infants with CHD (n = 10),^{14,27,29,30,32,35} in which the ICER was between $9837 and $139 051 per QALY; and infants with BPD or CLD (n = 6),^{27,29,32,35,47} in which the ICER was between $3984 and $40 036 per QALY. At a threshold of $100 000 per QALY, 86% of estimates for preterm infants, 86% of estimates for preterm infants with risk factors, 90% of estimates for infants with CHD, and 100% of estimates for infants with BPD or CLD were considered cost-effective. Other risk factors considered in preterm infants included chronological age at the beginning of the RSV season, school-aged siblings, day care attendance, smoking during pregnancy, male sex, and cystic fibrosis (in term infants only).^{25,37,39,42,52} From a societal perspective, palivizumab prophylaxis was considered a dominant strategy (ie, the strategy provided additional clinical benefit and was cost saving) in some instances for preterm infants,^28,42,52 term infants (with and without other risk factors),⁴⁰ and infants with CHD.²⁹ 

There were 5 studies in which cost-effectiveness was reported in cost per HA,^{14,24,33,34,48} of which 2 were industry funded.^24,34 In the study by Banerji et al,²⁴ the authors studied healthy term infants from a payer perspective in different regions of the Canadian Arctic and compared 2 scenarios of palivizumab prophylaxis for infants who were <6 months of age. The ICER for palivizumab prophylaxis ranged from being dominant (in specific Arctic regions) to $479 242 per HA in the Northwest Territories.²⁴ Also from the payer perspective, Hampp et al⁴⁸ assessed cost-effectiveness in preterm infants (<32 wGA) and term infants with CHD, CLD, and combinations of all 3 risk factors in a Florida setting. The ICERs were between $339 852 (preterm infants) and $2 406 129 per HA (healthy term infants without CLD or CHD).⁴⁸ 

From a societal perspective, the study by Rietveld et al³³ in 2010 from southwest Netherlands studied preterm infants (<28 wGA) with additional risk factors (male sex, birth weight <2500 g, and BPD). The ICER ranged between $21 066 and $1 331 529 per HA depending on the month of the prophylaxis. The most cost-effective month for palivizumab prophylaxis (lowest ICER) was December, whereas the least cost-effective month was October. In this study, the authors recommended a restricted immunization policy on the basis of their results.³³ Roeckl-Wiedmann et al³⁴ conducted a study in 2003 from southern Germany on preterm infants (<35 wGA) with additional risk factors. ICERs ranged between $10 011 and $308 658 per HA for preterm infants with CLD and preterm infants with risk factors (male sex, no CLD, and no siblings in school), respectively. In this study, the authors also recommended a restricted use of palivizumab in preterm infants with CLD.³⁴ 

Authors of 4 studies reported cost-effectiveness of palivizumab prophylaxis in other units: cost to prevent 1 day of hospitalization,⁴⁹ cost per LYG,⁵⁰ cost per case averted,⁴¹ and cost per RSV infection episode avoided.⁴⁵ In 3 of 4 studies, authors conducted analyses from a societal perspective.^41,49,50 Harris et al⁴⁹ conducted an economic evaluation on term infants with CHD in western Canada. The base-case ICER was $15 111 per 1 day of hospitalization prevented.⁴⁹ Hascoet et al⁵⁰ studied preterm infants (<32 wGA) with BPD or significant CHD in France. The base-case ICER was $36 971 per LYG and $28 198 per LYG for preterm infants with BPD and preterm infants with cardiopathy (CHD), respectively. The authors of this study used a cost-effectiveness threshold (unadjusted) of 45 000 Euros per LYG and considered prophylaxis cost-effective for both subgroups in France.⁵⁰ In New Zealand, Vogel et al⁴¹ studied preterm infants (<28 and 29–31 wGA) and infants with CLD. The ICER ranged between $28 265 per case avoided for preterm infants discharged from the hospital on oxygen and $164 176 per case avoided for preterm (29–31 wGA) infants with CLD. The authors concluded that the intervention was more cost-effective for preterm infants discharged from the hospital on oxygen followed by preterm infants of ≤28 weeks’ gestation.⁴¹ Lastly, Lofland et al⁴⁵ studied preterm infants with CLD in the United States. Their model used a reduction in incidence of RSV infection instead of a hospitalization reduction approach, ranging from a 50% ($56 313 per RSV infection episode avoided) to 83% reduction, in which palivizumab prophylaxis was considered a dominant strategy (ie, cost savings).⁴⁵ 

The cost-effectiveness of palivizumab prophylaxis compared with no palivizumab prophylaxis ranged from being a dominant strategy to $2 526 203 per QALY in preterm infants. Because studies estimated cost-effectiveness for varying ranges of wGA, we were unable to group all estimates into predefined intervals. For example, we did not group <29 wGA estimates under the <32 wGA estimates because we could not infer or reasonably assume the breakdown of the wGA in each preterm group. From the payer perspective, the ICER for palivizumab prophylaxis for infants born at <29 wGA (n = 3) ranged between $5188 and $20 038 per QALY.^35,36 For infants born at 29 to 32 wGA, the ICER (n = 3) ranged between $8337 and $48 430 per QALY.^35,36 At <32 wGA and <33 wGA, 2 estimates ($10 715–$21 130 per QALY)²⁸ and 3 estimates ($13 679–$35 151 per QALY) were identified, respectively.^32,47 In the 32- to 35-wGA range (includes 2 estimates at 32–35 wGA and 4 estimates at 33–35 wGA), there were 6 ICER estimates for preterm infants ($21 783–$756 060 per QALY)^{32,35,37,39,47} and 14 ICER estimates for preterm infants with additional risk factors ($177–$169 103 per QALY).^37,39 For preterm infants born at <35 wGA, there were 5 estimates between $25 838 and $791 265 per QALY.^14,27,29,32 All preterm infants with BPD or CLD were estimated in this <35-wGA group, in which the 4 estimates were between $12 653 and $111 171 per QALY.^14,27,29,47 

From a societal perspective, estimates of preterm infants born at 26 to 28 wGA were entirely extracted from Elhassan et al,⁴⁴ with ICERs between $140 341 and $2 043 230 per QALY. For preterm infants born at <29 wGA, there were 5 ICER estimates between $19 000 and $1 134 793 per QALY.²⁶ The cost-effectiveness of palivizumab from a societal perspective varied across studies for preterm infants born between 29 and 35 wGA, with ICER estimates between $381 427 and $920 300 per QALY (29–30 wGA),⁴⁴ being a dominant strategy (ie, cost savings and provides clinical benefits for <32 wGA),^28,42,52 $26 424 and $272 654 per QALY (32–35 wGA),^37,46 and $21 124 and $828 728 per QALY (<35 wGA).^14,32 In preterm infants (<35 wGA) with lung complications, 3 studies reported separate ICER estimates between $15 779 and $116 573 per QALY.^14,27,29 Six estimates were reported for preterm infants with risk factors between $18 401 and $522 514 per QALY ($18 401–$50 369 per QALY for two 32–34-wGA estimates and $43 023–$522 514 per QALY for four 32–35-wGA estimates).^42,52 

We stratified and plotted the ICERs for palivizumab prophylaxis expressed in cost per QALY in preterm infants by wGA in Fig 3; we presented 57 of 72 ICER estimates, stratified by study perspective, that were estimated <$200 000 per QALY.^† In Fig 3, 51 of the 57 (89%) ICER estimates for preterm infants (with or without other RSV risk factors) were below the $100 000-per-QALY threshold. Of the 15 ICER estimates excluded from Fig 3, 8 (ie, more than half) were from a single study by Elhassan et al,⁴⁴ whereas the rest were single estimates from other studies.^{14,26,35,39,42,46,52} 

Authors of 2 studies investigated the cost-effectiveness of palivizumab prophylaxis in Arctic (remote) regions of Canada.^24,40 In the study by Tam et al,⁴⁰ palivizumab was cost-effective from a health care–payer perspective for all infants from Baffin Island who were <1 year of age ($38 414 per QALY) or <6 months of age ($9926 per QALY), infants <1 year of age and from high risk areas for RSV ($325 per QALY), infants <1 year of age from remote areas ($24 109 per QALY), infants <6 months of age from remote areas (dominant), and infants <6 months of age from high risk areas for RSV (dominant). However, when compared with a $100 000-per-QALY threshold, it was not cost-effective for infants <6 months or for infants <1 year of age residing in Iqaluit.⁴⁰ Similarly, Banerji et al²⁴ concluded that their proposed palivizumab programs would be cost-effective in some but not all Arctic regions. The authors of both studies attributed the likelihood of these results to the higher hospitalization rates and transportations costs associated with hospitalization from remote areas to hospitals.^24,40 

Reduction in RSV hospitalization used in models ranged between 39% for infants with CLD in the United Kingdom³⁵ and 96% in healthy infants in a Canadian Arctic setting.²⁴ Mortality was reported in 19 studies, ranging between 1%^40,50 and 8.11%³² for various infant populations. The number of palivizumab doses per season was between an average of 3.88 doses in a 5-month season in Spain³⁷ and 6 doses in a 6-month RSV season.²⁴ Authors of most studies evaluated cost-effectiveness assuming 5 palivizumab doses per RSV seasons (n = 17), whereas the authors of 3 studies did not report the dose schedule.^35,46,48 The cost of a 100-mg vial of palivizumab in 2017 USDs was between $904 (from a UK study)³⁵ and $1866 (from a US study).⁴⁴ 

The most influential parameters reported across the 28 studies were the following: RSV hospitalization rates (43%),^‡ cost of palivizumab (36%),^§ discount rate (32%),^{26,–30,32,35,36,38} and efficacy of palivizumab (29%).^{33,–35,41,42,46,48,52} Other parameters that were influential in multiple studies included the following: mortality rate reduction, incidence of RSV (and/or sequelae), drug wastage resulting from vial usage, utility values (quality of life), and dosage scheme.

In our systematic review, we identified 28 relevant economic evaluations from OECD countries assessing the cost-effectiveness of palivizumab prophylaxis compared with no prophylaxis. The greatest number of cost-effectiveness estimates came from preterm infants, which was expected given their higher risk for RSV.^6,8 The majority of estimates for infants with or without additional risk factors (eg, BPD and CHD) were below the $100 000-per-QALY threshold. The only exception to this was the estimates for preterm infants from the societal perspective, in which only 48% of the estimates were <$100 000 per QALY. This exception was likely a result of a large group of estimates (n = 8; 35% of subgroup) extracted from 1 study, with estimates between $295 287 and $2 526 203 per QALY.⁴⁴ Possible reasons for the higher ICERs in this study may include: using the highest adjusted cost for a 100-mg vial of palivizumab at $1866, following infants only up to 8 years of age, and, as suggested by the authors, overestimating the quality of life (utility) for subsequent asthma onset. Separate sensitivity analyses reducing the palivizumab cost by 25% and reducing the health-state utility value of asthma afforded ICERs <$200 000 per QALY and <$100 000 per QALY, respectively.⁴⁴ 

On the basis of our review, the cost-effectiveness of palivizumab prophylaxis varies depending on the population and setting. To facilitate comparisons and summarize our findings, we adjusted all ICERs to 2017 USDs per QALY and stratified them on the basis of gestational age at birth and risk factors for RSV in Fig 3. For term and preterm infants with BPD or CLD, the ICER was <$50 000 per QALY in 9 of the 10 estimates from a payer perspective. All other subgroups of infants (term, preterm, and with CHD or other risk factors) resulted in inconsistent results for palivizumab prophylaxis, with the intervention being dominant at times and having an ICER up to $791 265 per QALY in other scenarios. When stratifying for preterm births by wGA, we noticed lacking evidence for infants born at <28 wGA, especially from the payer perspective. No specific trend was depicted between the wGA and the ICER, overall or stratified by perspective. However, we should note that although preterm estimates were available across 26 to 35 wGA, estimates for preterm infants with additional risk factors or BPD or CLD were limited to 33 to 35 wGA. Generally, one would expect ICERs from a societal perspective to be lower than those from a payer perspective, but on the basis of our review, this trend does not exist for 2 potential reasons: (1) payer and societal perspective estimates were coming from different studies and (2) there was heterogeneity in model designs and differences between setting-specific costs and RSV epidemiology.

Because palivizumab prophylaxis was determined to be cost-effective in some settings but not cost-effective in others, we summarized the most frequently reported influential parameters affecting the ICER, which were the RSV hospitalization rates and cost of palivizumab used. Reduction in RSV hospitalization varied drastically between 39% and 96% depending on the population of interest and the source of the data. The cost of a 100-mg vial of palivizumab also ranged between $904 and $1866 (2017 USDs). The influential nature of both parameters was expected given that reduction in RSV and RSV hospitalization is essential to reduction in costs and future sequelae, whereas the costs of palivizumab is directly related to the ICER. However, it was interesting to note that vial usage and dosage scheme only affected the ICER in 4^27,28,38,41 and 3 studies,^37,41,47 respectively. In studies where drug wastage through vial usage was addressed, ICERs fluctuated up to 50% depending on the assumed vial usage. In a New Zealand study, assuming no vial sharing (the entire 100-mg vial is used per injection) increased costs up to 50%,⁴¹ whereas authors of a study from Spain reported a lower ICER when 50-mg vials were used instead of 100-mg vials.²⁸ It has been suggested in the literature and by physicians that vial-usage efficiency can be achieved for palivizumab.⁵³ Authors of many studies did not assess scenarios in which the vial usage became more efficient or the number of assumed doses was reduced, which remains a question that can be addressed in future studies.

The cost-effectiveness of palivizumab prophylaxis has been explored in multiple reviews in the past 2 decades,^11,12,14 but only 4 have been published between 2010 and 2013.^15,54,–56 Our results and conclusions are consistent with those in other reviews and are most comparable with the systematic review by Smart et al⁵⁴ published in 2010 in which the authors reported ICERs (in 2009 Canadian dollars) for palivizumab prophylaxis varying between being dominant and $3 365 768 per QALY depending on the study population, outcomes, and model parameters. We added onto this review by capturing studies from 2010 to mid-2018 but limited our scope to OECD countries and adjusted for inflation differences by using the purchasing power parity rates from the OECD. In their reviews, Andabaka et al¹⁵ and Prescott et al⁵⁶ similarly concluded that cost-effectiveness of palivizumab was inconsistent. Hussman et al⁵⁵ conducted a review on RSV prophylaxis overall and included studies in which palivizumab and other interventions (eg, respiratory syncytial virus immune globulin intravenous) were compared. To our knowledge, our review is the first to provide an update on the cost-effectiveness of palivizumab prophylaxis compared with no prophylaxis since the 2014 American Academy of Pediatrics (AAP) guideline update.⁵⁷ 

Our review has several limitations. Differences in model designs, RSV hospitalization rates used, disease progress, study perspectives, and settings prevented us from providing definitive conclusions on the value of this intervention. We attempted to summarize the cost-effectiveness of this intervention from 2000 to 2018 but acknowledge that changes in AAP recommendations in the United States (and decision-makers in other respective countries) over time can affect model design and input data. Lastly, our review may be subject to publication and language bias because we did not search the gray literature or include articles not in English or French.

Despite these limitations, with our review, we provide a comprehensive summary of the cost-effectiveness of palivizumab prophylaxis from OECD countries to inform decision-makers of the estimated value of this intervention in term infants, preterm infants, and infants at high risk for RSV (eg, CHD and BPD or CLD). We extracted all base-case results and scenario analyses to create Fig 3, which gives a sense of the number of studies (and estimates) that fall under specific cost-effectiveness thresholds from both payer and societal perspectives. We standardized all estimates to 2017 USDs, which allowed us to group, stratify, and compare the cost-effectiveness estimates in costs per QALY. These adjusted ICERs should be useful for program decision-makers because costs can be significantly underestimated if not appropriately inflated.

Palivizumab prophylaxis for RSV can be considered cost-effective in certain subgroups of infants according to predefined cost-effectiveness thresholds but varied depending on study setting, population of interest, risk factors, and input parameters. From a payer perspective, palivizumab was found to be relatively cost-effective in infants with BPD or CLD, infants with CHD, term infants from specific remote communities, and preterm infants with and without lung complications. Authors of future studies should take into account all influential parameters presented in this review, especially concerns regarding vial usage and dosage because these can drastically reduce the costs associated with the intervention and impact the model outcomes.

We thank Public Health Agency of Canada librarian Lynda Gamble for her assistance in developing the search strategy and managing results.

BPD

bronchopulmonary dysplasia

CHD

congenital heart disease

CLD

chronic lung disease

HA

hospitalization averted

ICER

incremental cost-effectiveness ratio

JBI

Joanna Briggs Institute

LYG

life-year gained

NHS

National Health Service

OECD

Organization for Economic Cooperation and Development

QALY

quality-adjusted life-year

RSV

respiratory syncytial virus

USD

US dollar

wGA

weeks’ gestational age