Model Structure

The model employs a cohort framework and Markov-type process to depict clinical outcomes and economic costs of RSV-positive respiratory tract illness (RTI) from birth to age 11 months, lifetime consequences of premature RSV-related death, and the expected impact of maternal vaccination with RSVpreF vaccine on the aforementioned outcomes among infants (Fig. 1). The model was adapted from a model presented at the ACIP meeting in September 2023 [20] and was calibrated to reflect Japanese healthcare settings.

Fig. 1

Decision tree structure for cost-effectiveness analysis. In the cost-effectiveness analysis, a new combination RSV prophylaxis was compared with the current prophylaxis by epidemiology, costs, and cost-effectiveness. Neonates were divided into subgroups by their mothers’ vaccination status and subsequent usage of prophylaxis. RSV incidence, usage of medical resources, and subsequent mortality and productivity losses were calculated for each of the subgroups. RSV respiratory syncytial virus

Clinical outcomes included RSV cases, RSV-related deaths, life years, and quality-adjusted life years (QALYs) for each healthcare setting including hospitalization, outpatient visits, and emergency department (ED) visits. Economic outcomes included intervention costs (costs for RSVpreF vaccine and palivizumab), medication costs for infants, and productivity losses. The model cohort was a monthly rolling cohort of infants aged less than 1 year. The analysis included 12-monthly cohorts; for each cohort, RSV cases, deaths, and costs are tallied only during the first year of life. The gestational age of infants was considered and divided into four age categories: full-term (≥ 37 weeks gestational age, wGA), and preterm (≤ 27 wGA, 28–31 wGA, and 32–36 wGA) [21]. In order to maintain internal consistency between international resources we used the definition of World Health Organization (WHO) for term pregnancy rather than the US definition [22, 23]. In the simulation, the number of cases was estimated for each healthcare setting on the basis of the epidemiologic inputs. Subsequent RSV-related deaths were calculated by multiplying the estimated number of cases in each care setting by case fatality rates. Life years were computed by aggregating life years for the first year of life and full life expectancies for the remaining alive population from all birth categories. Deaths of infants aged < 1 year were counted as 0.5 year. Along with the clinical and economic outcomes, incremental costs, incremental QALY, and incremental cost-effectiveness ratios (ICERs) per QALY gained were estimated from a payer perspective and a societal perspective, assuming a cost-effectiveness threshold of ¥5 million (US $38,052) per QALY gained. Model was adapted with a lifetime time horizon and an annual discount rate of 2.0% [24].

Intervention Strategies

As of February 2024, Japanese clinical guidelines for RSV prophylaxis did not yet include the addition of RSVpreF vaccination to the existing prophylaxis, palivizumab [9]. We assumed that RSVpreF vaccine and palivizumab will be used together in the clinical practice as a comprehensive prophylaxis program to improve protection against infants from RSV infection. At present, palivizumab is approved and highly used for premature who were born  ≤ 35 wGA and infants with risks of RSV disease, such as infants with bronchopulmonary dysplasia (BPD), congenital heart disease (CHD), Down’s syndrome, or a compromised immune system in Japan [11, 12, 14, 25]. In contrast, we assumed that RSVpreF vaccination is effective for full-term (≥ 37 wGA) and preterm born at 32–36 wGA, and some proportion of infants that could be protected from maternal immunization could be exempted from palivizumab prescription. In this study, two cases, scenario 1 (base case scenario) and scenario 2 were developed by changing the target population for palivizumab prescription in the combination prophylaxis (Table 1).

Table 1 Current and combination prophylaxis strategies of RSV infection

For both cases, all women were eligible for year-round RSVpreF vaccination. In scenario 1, premature (≤ 31 wGA) without risk and all infants with risk were eligible to receive palivizumab. In this setting, premature (32–36 wGA) without risk was assumed not to receive palivizumab if their mother received RSVpreF vaccine. However, if their mothers were unvaccinated or infants born within 2 weeks after maternal vaccination, we assumed that the people were not sufficiently protected by transferred maternal antibodies and received palivizumab. In scenario 2, premature (≤ 31 wGA) regardless risk in infants were eligible to receive palivizumab. These combination prophylaxes were compared with the current prophylaxis (palivizumab alone) in the analysis (Table 1).

Model Input Parameters

The input parameters for the basic analysis are listed in Table S1.

Population Inputs

Values for birth-related parameters such as number of infants born during the 12-month period were extracted from the Vital Statistics Survey of Japan in 2021 and further calibrated to fit into the model [26]. The percentage of infants with high risk was assumed to be 2.58% based on the analysis of a Japanese claims database [15].

Epidemiological Inputs

The annual rates of RSV cases were extracted from the Japanese claims database analysis of 2021 and adjusted to mimic the condition where palivizumab was not yet utilized in Japan (Table S2) [7]. The proportion of RSV with lower respiratory tract illnesses (LRTI) in hospitalized cases was 91.6% [27]. Relative risk of RSV encounters was estimated from data obtained in the USA because of unavailability of relevant data in Japan [28]. The probabilities of being infected by RSV in each month were estimated from an average of the distributions of RSV encounters between 2017 and 2019 because the peak season of RSV was different in each year [29].

Mortality Inputs

Among mortality inputs, infant mortality rate and relative risk of infant mortality were based on the Vital Statistics Survey of Japan in 2021 [26]. Case fatality rate was calculated from the cases reported in 2020 [30]. As a result of lack of Japanese data, relative risk of death due to RSV was based on data from the literature and statistics from the Centers for Disease Control and Prevention [2, 31].

Intervention-Related InputsMaternal Vaccine

RSVpreF vaccination rate was set as 80% in scenario 1 based on the maternal vaccination rate against COVID-19 in Japan [32]. Vaccine effectiveness (VE) was estimated using the cumulative efficacy data of the pivotal phase 3 clinical trial (ClinicalTrials.gov Identifier NCT04424316) [33]. After 180 days, the VE was assumed to decrease linearly in the next 3 months until effectiveness rate reaches 0%. The VE values of RSV-positive LRTI requiring hospitalization and RSV-RTI (which includes RSV-LRTI and RSV-upper respiratory tract illnesses) for ED and outpatient settings were used for model input (Table 1). The aforementioned study was not powered to provide estimates of VE among preterm infants; therefore, the VE for preterm (32–36 wGA) was assumed to be 83.3% of the corresponding values for full-term infants (≥ 37 wGA) based on an ongoing observational sero-epidemiology study of naturally acquired RSV antibody transplacental transfer [34]. In the study, geometric mean transplacental transfer cord/maternal neutralizing antibody titer ratio for RSV A/B was reported to be 1.2 among full-term infants (≥ 37 wGA) and 1.0 among infants born at 32–36 wGA (1.0/1.2 = 83.3%). The VE values among other preterm (≤ 31 wGA) infants with risk and infants born < 2 weeks after maternal administration of RSVpreF vaccine irrespective of term status were assumed as 0% as a result of lack of evidence.

Palivizumab

Prescription rate of palivizumab among preterm infants was assumed to be 0% for 36 wGA and 90% for ≤ 35 wGA based on the literature and pediatric expert opinion [11, 12]. Thus, the weighted average 46.2% was applied for prescription rate among the preterm population (≤ 36 wGA). For the high-risk population, 95% was applied as a high prescription rate among this population was previously reported [11, 12, 25]. Palivizumab was administered six times during the peak season of RSV on the basis of the analysis of Japanese claims database [15]. Effectiveness of palivizumab was set as 56% based on systematic review [9].

Cost InputsDirect Costs

Direct costs included the costs of intervention (vaccine/drug costs with administration costs) and medical care. Administration costs were based on the national medical service fee points set by the Ministry of Health, Labour and Welfare [35]. As the unit cost of RSVpreF vaccine was not yet determined in Japan, a range of vaccine prices were explored from ¥5000 to ¥50,000 (US $38 to US $380). Drug prices of palivizumab per dose were obtained by dividing the mean cumulative drug costs of palivizumab by the number of admissions [15]. Medical care costs per episode in three different medical care settings were obtained through Japanese claims database analysis in 2022 [15].

Productivity Loss

Three potential sources of productivity losses were considered in the analysis: (1) the caregivers’ absence from work due to providing care to infected infants, (2) lost labor opportunities that might have been gained after maturity of the children who died of RSV infection, and (3) absence from work of pregnant women receiving maternal vaccination at a hospital visit and caregivers taking children to hospital for palivizumab administration. These productivity losses were calculated by multiplying the average daily wage and the number of working days missed. We assumed that all caregivers were in full-time employment and productivity losses due to caregiving tasks were incurred by all infants on the basis of input from a health economics expert. The number of days of parental care in ED and outpatient settings were determined on the basis of pediatric expert opinion. For hospitalized population, we added 3 days of home care in addition to average hospitalization period of 7 days [36]. For future productivity loss we assumed that a fully matured adult will be in the workforce between 18 and 64 years old and receive the average daily wage. The number of workdays lost as a result of maternal vaccination or taking infants to receive palivizumab administration was assumed 0.5 days per dose, based on input from health economics expert. As a result of lack of reference data, we assumed that productivity loss will arise in 50% of the aforementioned applicable population as no additional productivity losses shall be incurred if maternal vaccination or palivizumab administration were part of regular checkups on the basis of the discussion with an expert in health economics.

All costs were converted to US dollars using the exchange rate for 2022 provided by the Organisation for Economic Co-operation and Development (OECD; US $1 = ¥131.4) [37].

Health Utility Inputs

Baseline utility without RSV infection was set as 1.0 among infants aged 0–11 months. QALY loss due to RSV infection in each medical setting was extracted from a previous study [38]. The general population utility was set as 1.0 for 1–17-year-olds on the basis of input from a health economics expert. Utility for 18–99-year-olds was calculated on the basis of the population data of 2021 using the EuroQol 5-dimension 5-level (EQ-5D-5L) scores [39, 40]. These inputs were used to calculate future lost QALYs due to premature RSV-related mortality.

Analysis

Basic analysis for two cases was performed to analyze the impact of varying vaccine prices on ICER. One-way deterministic sensitivity analysis was performed to assess the impact of each parameter and identify key drivers for ICER other than vaccine price. The model parameters were examined at a lower and upper bound of ± 25%. For other scenarios of the sensitivity analyses, the input conditions are summarized in Table S3.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.