Plan of Analysis, Design, and Setting

This study evaluated the public health impact and economic burden of RSV in Colombia using a static cohort decision model. The analysis compared SoC with nirsevimab and RSVpreF and additionally performed an indirect comparison between nirsevimab and RSVpreF to inform national RSV prevention decisions. The model estimated RSV-related healthcare resource utilization, mortality, QALYs, and costs in infants aged < 1 year.

The study was conducted in the Colombian healthcare system. Epidemiologic and healthcare resource utilization inputs were extracted from the National Institute of Health (from the Spanish acronym—INS, Supplementary Material S1) sentinel surveillance (predominantly pediatric hospital and outpatient settings) and the Colombia’s Health Services Utilization Database (from the Spanish acronym SISPRO), complemented by a targeted literature review of Colombian studies [3, 4, 13] and expert validation. Model structure and reporting followed ISPOR good research practices and CHEERS 2022 guidance (Fig. 1) [14, 15]. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Fig. 1

Model structure. ER emergency room, ICU intensive care unit, LRTI lower respiratory tract infection, MA medically attended, MV mechanical ventilation, PC primary care, RSV respiratory syncytial virus

Target Population

We analyzed the newborn cohort in Colombia at risk of their first RSV season using vital statistics for 2023 [16]. The monthly infant at-risk was estimated by considering the monthly distribution of births from the average of the last 5 years. Infants were divided into three groups according to the risk of developing LRTI: (1) palivizumab-eligible (< 32 weeks’ GA or with bronchopulmonary dysplasia (BPD), chronic lung disease, congenital heart disease, and/or neuromuscular disease), (6.54%, n = 33,366); (2) preterm < 35 weeks’ GA without bronchopulmonary dysplasia (BPD)/chronic lung disease/congenital heart disease, and/or neuromuscular disease) (1.38%, n = 7045); and (3) late preterm/term infants ≥ 35 weeks’ GA (92.08%, n = 469,947).

Intervention and Comparators

Two interventions (nirsevimab and RSVpreF) and one comparator (palivizumab SoC) were evaluated across five RSV prevention strategies (Sect. “Description of Immunization Strategies and Analytical Scenarios”). Nirsevimab was modeled as a single-dose monoclonal antibody administered to infants. RSVpreF was modeled as maternal immunization between 32 0/7 and 36 6/7 weeks’ GA, providing passive protection to newborns. The comparator reflected current practice in Colombia using palivizumab as monthly prophylaxis in eligible high-risk infants (up to five doses) during periods of high RSV circulation [8] (product and administration details are presented in Supplementary Material S2).

Intervention Period

Colombia has year-round RSV circulation with seasonal peaks that vary by region and year [4]. Seasonality was parameterized using SISPRO (2015–2022) and cross-checked against INS surveillance data (Fig. 1S) [5, 17]. RSV was assumed to circulate year-round, with a major peak between March and July and a smaller peak around November–January. We defined the primary RSV season as March–July and evaluated two implementation windows: year-round (12 months; administration at birth) and seasonal (5 months; administration at birth for infants born during the season with catch-up in March for infants born outside the season). Additional details on seasonality triangulation and intervention window definitions are provided in Supplementary Material S3.

Rationale and Model Description

A static cohort decision analytic model (Global Sanofi) was used. This model evaluates the costs and clinical outcomes associated with nirsevimab and RSVpreF treatment in three groups of infants (palivizumab-eligible, preterm, and term) against RSV infections, covering a time horizon of 1 year, with long-term follow-up for sequelae and complications, and lifetime for years of life lost due to RSV-related premature death. The model includes different types of medical visits (primary, outpatient, and emergency) and hospitalizations for LRTI, differentiating between hospitalizations in regular wards, intensive care unit (ICU), and ICU with mechanical ventilation. It also considers complications arising from hospitalization, RSV-related mortality, and other causes of mortality. The framework allows for the evaluation of the economic and clinical impact of the intervention over different periods and in different patient groups.

Parameters and Characterization of RSV in Colombia

To parameterize RSV risk in infants aged < 1 year, we triangulated national surveillance (INS), administrative healthcare utilization data (SISPRO), and published evidence [5, 17]. INS data were used as the primary source for RSV positivity and incidence estimates, whereas SISPRO data were used to characterize the infant-specific monthly distribution of medically attended respiratory events. RSV positivity estimates across sources were 31.6% (INS 2023) and 26.87% (WHO 2017–2019 and 2022); the INS estimate was selected for the base case due to national coverage and surveillance robustness [5]. To account for potential under-ascertainment outside urban sentinel sites, we applied an adjustment informed by the rural population share (DANE 2022) (Supplementary Material S4) [18]. Full derivations for age- and month-specific RSV risk and related assumptions are provided in Supplementary Material (S4–S6), and parameter values are summarized in Table 1. For palivizumab-eligible infants, baseline hospitalization rates prior to palivizumab were derived via back-calculation using observed post-introduction rates, assumed coverage (69%), and palivizumab efficacy (51%) [19] (Supplementary Material S6). Recurrent wheezing probabilities over 3 years following RSV hospitalization were taken from the literature [20].

Efficacy and Coverage Assumptions

Efficacy inputs were obtained from a literature review and published clinical trials for each technology assessed. We assumed that nirsevimab was non-inferior to palivizumab in the palivizumab-eligible population. Palivizumab was modelled as providing 51% protection against RSV outcomes in eligible infants [19], with protection assumed to last 30 days per dose [19, 21] and up to five monthly doses in line with standard practice. For nirsevimab, efficacy was modelled as constant (flat) over the 5-month protection window (150 days). Protection rate against inpatient outcomes was 83.2% for preterm, late-preterm, and term infants [24, 25], whereas in outpatient settings, this percentage was 86.2% for preterm infants and 74.5% for late-preterm and term infants [11, 26, 27].

Maternal RSVpreF vaccination was modelled as reducing infant RSV hospitalizations by 56.8% within the first 6 months of life and outpatient events by 51.3%. RSVpreF efficacy was modelled from birth to 180 days with waning over time [12, 28, 29]. RSVpreF time-varying efficacy was modelled using a polynomial function fitted to published efficacy timepoints (as reported in VRBPAC monthly summaries and MATISSE outputs), allowing interpolation between observed values.

Coverage assumptions were based on national immunization programme data and calibration to observed utilization: nirsevimab coverage at birth was assumed to be 93.5% [30], maternal RSVpreF coverage 74%, and palivizumab coverage 62.77% based on annual sales. More details on efficacy and coverage assumptions are described in Supplementary Material S7.

Perspective

Two key perspectives on the public health and economic impact of RSV were evaluated. From the payer’s perspective, we considered direct event-related medical costs and quality-adjusted life years (QALYs) in infants. From a societal perspective, we additionally included direct non-medical costs (out-of-pocket expenditures), indirect costs (e.g., caregiver productivity losses), and QALYs in both infants and caregivers, providing a broader view of the public health and economic impact of RSV prevention strategies.

Description of Immunization Strategies and Analytical Scenarios

The evaluated strategies were (i) Immunization with palivizumab, the SoC currently recommended for high-risk infants; (ii) Nirsevimab: Passive immunization of newborns entering the RSV season with an all-year-round immunization strategy; (iii) Immunizations with RSVpreF targeted pregnant women between 32 0/7 and 36 6/7 weeks’ GA. Results were obtained by evaluating the nirsevimab and maternal RSVpreF against SoC across five immunization strategy scenarios (S1–S5) that combined perspective (payer vs societal) and delivery approach (year-round vs seasonal with catch-up). RSVpreF was consistently modeled as a year-round strategy for pregnant women (32 0/7 to 36 6/7 weeks’ gestational age). From a payer perspective, the model evaluated year-round infant administration (S1), seasonal infant administration (S2), and a year-round parity scenario assuming equal coverage for RSVpreF and nirsevimab interventions (S5). From a societal perspective, the model evaluated year-round (S3) and seasonal (S4) infant administration. Full scenario definitions are provided in the Supplementary Material S8.

Time Horizon

The time horizon of the burden analysis of nirsevimab in infants with RSV in Colombia was 1 year. The horizon can be adapted to short- or long-term goals and can even extend to the patient’s lifetime, considering all-cause mortality and the long-term impacts of RSV infection.

Discount Rate

In the base case of the analysis, the results are presented both without discounting and applying a discount rate of 5% to health costs and benefits. This rate was considered the most appropriate for the Colombian context, in line with the recommendations of the Institute for Health Technology Assessment (IETS) established in its methodological manual. The choice of 5% reflects the need to capture the present value of health costs and outcomes over time.

Outcome Selection

The effects of RSV on disease burden were measured by considering various events, as previously mentioned. Regarding complications of RSV-related events in the long term, recurrent wheezing was the only one captured in the model, and complications were simulated as a one-time risk over a 3-year time horizon. In addition, a comprehensive analysis of disease-related costs was conducted, including a detailed assessment of expenditures and QALYs associated with various RSV-related health outcomes. The QALY loss per event was obtained from the RESCEU prospective study on healthy term infants [31], utilities associated with recurrent wheezing were obtained from Li et al. 2022 [20], and the QALY loss due to RSV-related premature death was estimated, based on the remaining life expectancy at the average age of death from RSV, adjusted by an age-specific utility weight and discounted at the annual rate.

To evaluate the effectiveness of the proposed intervention strategies, we used the number needed to immunize (NNI), a public health metric indicating how many individuals must be immunized to prevent a single RSV-associated adverse event.

Outcome Measurement and Validation of Model Parameters

Outcome measurements in this economic analysis were performed following the principles of the Institute for Health Technology Assessment (IETS). Health outcomes were expressed in QALYs, and the direct and indirect costs of care were considered. The validation of the model parameters and assumptions was based on a targeted literature review in Colombia, complemented by the experience of local clinical and methodological experts, ensuring the relevance of the results for the national context.

Measurement and Assessment of Resources and Costs

Model inputs for healthcare resource utilization and costs are described in Table 2. Resource assessment was performed based on the following key cost categories: hospitalization, ICU, emergency department (ER) visits, mechanical ventilation, and primary care visits. In addition, direct non-medical costs were included from a societal perspective. ICU use, including hospitalization, was derived from previous local studies [13, 32], which were used to estimate the average cost per day of ICU stay. ER visits were calculated from specific cohort registries, in which the use of emergency services for patients with similar conditions was detailed, allowing a more granular approach to cost estimation. From a societal perspective, out-of-pocket expenditure was estimated with standard resources use and cost unit from Salcedo-Mejía et al. [32], indirect cost was obtained with minimum wage by law [33] in Colombia, and the productivity time loss from medical attention and premature deaths using human capital approach [34,35,36]. Total costs were calculated based solely on health event-related expenses. Acquisition costs or prices related to the prophylactic interventions were not included, as official negotiations or reference prices are not currently available for all evaluated technologies. All costs were reported in 2023 US dollars (TRM 3800 Colombian pesos) and adjusted for inflation when required. The average costs reflect the study period estimates for hospitalization and RSV care services.

Table 2 Model inputs for healthcare resource utilization and costsCharacterization of Uncertainty

Uncertainty related to the parameters affecting estimates of RSV infection risk, hospitalization costs, and effectiveness of interventions was addressed through deterministic (DSA) and probabilistic (PSA) sensitivity analyses. To obtain more precise estimates, outliers derived from the pandemic were excluded by using the most representative scenario. This robust approach allowed us to characterize the sources of uncertainty in the analysis and improve our understanding of the variability in the results.

Modeled Population and Risks

The study population consisted of a simulated cohort of 510,357 live births in Colombia (2023), followed from birth to 12 months of age and stratified into the three risk groups described in Sect. “Target Population” and summarized in Table 1. Baseline RSV hospitalization rates in the first year of life were 10.99% for palivizumab-eligible infants, 5.96% for preterm infants, and 3.70% for late preterm/term infants; ICU admission parameters are summarized in Table 1 (including 47.5% among late preterm infants and 14.53% among term births). ER and primary care visits were estimated by applying nationally derived service-use ratios to age-specific RSV hospitalization incidence (ER-to-hospitalization ratio = 1.20; consultations-to-hospitalization ratio = 9.74). Additional details are provided in the Supplementary Material S9.