Respiratory syncytial virus (RSV) causes substantial hospitalization in US infants. The Advisory Committee on Immunization Practices recommended nirsevimab in infants younger than 8 months born during or entering their first RSV season and for children aged 8 to 19 months at increased risk of RSV hospitalization in their second season. This study's objective was to evaluate the cost-effectiveness of nirsevimab in all infants in their first RSV season and in high-risk children in their second season.
We simulated healthcare utilization and deaths from RSV with and without nirsevimab among infants aged 0 to 7 months and those 8 to 19 months old over a single RSV season. Data came from published literature, US Food and Drug Administration approval documents, and epidemiologic surveillance data. We evaluated societal outcomes over a lifetime discounting at 3% and reporting in 2022 US dollars. Sensitivity and scenario analyses identified influential variables.
We estimated that 107 253 outpatient visits, 38 204 emergency department visits, and 14 341 hospitalizations could be averted each year if half of the US birth cohort receives nirsevimab. This would cost $153 517 per quality-adjusted life year (QALY) saved. Nirsevimab in the second season for children facing a 10-fold higher risk of hospitalization would cost $308 468 per QALY saved. Sensitivity analyses showed RSV hospitalization costs, nirsevimab cost, and QALYs lost from RSV disease were the most influential parameters with cost-effectiveness ratios between cost-saving and $323 788 per QALY saved.
Nirsevimab for infants may be cost-effective, particularly among those with higher risks and costs of RSV.
Respiratory syncytial virus (RSV) causes substantial hospitalization in US infants and nirsevimab has been shown to reduce lower respiratory illness from RSV in infants and children.
Nirsevimab for infants may reduce 107 253 outpatient visits and 14 341 hospitalizations per year at a cost of $153 517 per quality-adjusted life year saved in the United States. Cost-effectiveness is influenced by hospitalization costs, nirsevimab cost, and RSV quality-of-life.
Respiratory syncytial virus (RSV) causes substantial morbidity in infants in the United States. It is estimated that hospitalization for RSV in infants costs the US society $472 million per year.1 The disease and economic burden of severe RSV is particularly high for infants in their first few months of life.2–4
In July 2023, the US Food and Drug Administration approved a monoclonal antibody, nirsevimab, for use in infants and young children to prevent RSV lower respiratory tract infection (LRTI). Nirsevimab is an immunoglobulin G1 antibody that binds to the prefusion conformation of the RSV F protein and was engineered to have a prolonged half-life.5 In clinical trials, nirsevimab effectiveness against medically attended (MA) RSV LRTI for up to 150 days was 79.0%; side effects were rare and included rash and injection site reactions.6 In August 2023, the Advisory Committee on Immunization Practices (ACIP) recommended nirsevimab for all infants aged <8 months born during or entering their first RSV season and for infants and children aged 8 to 19 months who are at increased risk for severe RSV disease and entering their second RSV season.6 Before approval, palivizumab was the only available monoclonal antibody immunization, but it requires multiple costly doses, and has been only recommended to be given during the RSV season for a very small subset of infants who are high-risk premature or those born with certain heart or lung conditions. Because the duration of protection is limited and RSV is highly seasonal,7 nirsevimab is intended to be administered as a single dose at birth for infants born during the RSV season or shortly before the RSV season begins, typically in October or November, for children born outside of the RSV season to provide these children protection during their first RSV season.6 In September 2023, ACIP recommended a maternal RSV vaccine for use during pregnancy to prevent infant LRTI. Either use of the maternal RSV vaccine or infant receipt of nirsevimab was recommended, but both are not needed for most infants.8
The ACIP incorporates cost-effectiveness information into its decision-making process when developing immunization recommendations. The findings of our analysis were presented to ACIP as part of its deliberations on nirsevimab recommendations.9
The objective of this analysis was to evaluate the cost-effectiveness of nirsevimab for infants in their first RSV season. In a secondary analysis, we evaluated the cost-effectiveness of nirsevimab administration to high-risk young children in their second RSV season. In an additional secondary analysis, we evaluated the use of nirsevimab in infants born to mothers who had been vaccinated with the maternal RSV vaccine. Overall, we aim to provide further transparency on the methods and results of our cost-effectiveness study used by ACIP as part of its deliberations leading to the 2023 policy recommendation on nirsevimab. A separate article evaluates maternal vaccination.10
Methods
To evaluate the projected health benefits and cost-effectiveness of nirsevimab, we did not collect direct patient data, but instead created a decision analytical model using secondary data that simulates the short- and long-term impacts of RSV on infants with and without nirsevimab. For the base case scenario with nirsevimab, we assumed 50% uptake of nirsevimab, administered at birth for those born between October 1 and March 31 and administered in October (for those born in April, June, and August) or in November (for those born in May, July, or September) to correspond with routine well-child visits at about 2, 4, and 6 months of age. Since we do not simulate transmission, the resulting value per person immunized is not affected by uptake. RSV disease and economic outcomes were estimated using a societal perspective in the base case.
Model Description
Assuming a uniform monthly distribution of US births in an annual birth cohort, the model simulates the number of births throughout each month over a 1-year time frame and accounts for lifetime outcomes (eg, lives and life-years lost because of RSV). The cohort of infants would face variable risk of RSV MA-LRTI on the basis of the specific month of birth and RSV seasonality (Supplemental Fig 4). The model incorporated RSV-associated disease outcomes along with resource utilization such as outpatient visits, emergency department (ED) visits, hospitalizations, and deaths (Fig 1). Likewise, we included adverse events from nirsevimab such as injection site reactions, systemic reactions, or theoretical serious adverse events. Each disease and adverse event outcome was associated with costs and health-related quality-of-life losses.
Epidemiology Inputs
Annual RSV incidence was estimated from a variety of epidemiologic studies of RSV (Table 1). RSV inpatient incidence came from 2016 to 2020 data from the Centers for Disease Control and Prevention’s New Vaccine Surveillance Network.4 We assumed 100% of RSV hospitalizations were because of LRTI.11 ED and outpatient incidence came from several New Vaccine Surveillance Network epidemiologic studies in young children and infants.2,3,12 Because these were less severe outcomes, a fraction of those events were assumed to be because of LRTI (whereas the remaining were assumed to be because of upper respiratory tract infections [URTIs]).11 Death could happen for those hospitalized with RSV, with mortality rates based on data among all infants hospitalized with RSV (both high risk and low risk).12–15 We did not incorporate potential long-term sequelae of RSV infection (eg, asthma). Monthly RSV seasonality was derived from prepandemic (2015–2019) proportion of RSV detections per month from the National Respiratory and Enteric Virus Surveillance System16 (Supplemental Fig 4).
Model Parameters
Model Input Parameter . | Base case . | Range . | Distribution . | Source . |
---|---|---|---|---|
Category of RSV incidence per 100 000 | ||||
Inpatient | ||||
Age 0 mo | 1760 | 1560–1970 | Lognormal | CDC NVSN4 ,a |
Age 1 mo | 3110 | 2850–3390 | Lognormal | |
Age 2 mo | 2230 | 2030–2450 | Lognormal | |
Age 3 mo | 1560 | 1390–1740 | Lognormal | |
Age 4 mo | 1360 | 1200–1520 | Lognormal | |
Age 5 mo | 1090 | 960–1250 | Lognormal | |
Age 6 mo | 960 | 810–1120 | Lognormal | |
Age 7 mo | 800 | 640–960 | Lognormal | |
Age 8 mo | 730 | 600–880 | Lognormal | |
Age 9 mo | 840 | 680–990 | Lognormal | |
Age 10 mo | 600 | 480–730 | Lognormal | |
Age 11 mo | 600 | 490–730 | Lognormal | |
Age 12 mo | 630 | 500–750 | Lognormal | |
Age 13 mo | 500 | 380–620 | Lognormal | |
Age 14 mo | 580 | 470–700 | Lognormal | |
Age 15 mo | 540 | 430–660 | Lognormal | |
Age 16 mo | 400 | 290–510 | Lognormal | |
Age 17 mo | 370 | 270–460 | Lognormal | |
Age 18 mo | 370 | 260–480 | Lognormal | |
Age 19 mo | 340 | 250–450 | Lognormal | |
Age 20 mo | 280 | 190–370 | Lognormal | |
Age 21 mo | 210 | 140–290 | Lognormal | |
Age 22 mo | 180 | 120–260 | Lognormal | |
Age 23 mo | 290 | 200–380 | Lognormal | |
Proportion with LRTI | ||||
Age 0–5 mo | 1 | 0.5–1.0 | — | Assumption based on Rainisch 202010 |
Age 6–23 mo | 1 | 0.5–1.0 | — | Assumption based on Rainisch 202010 |
ED | ||||
Age 0–5 mo | 7500 | 5500–7500 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)2 |
Age 6–11 mo | 5800 | 5700–5800 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)2 |
Age 12–23 mo | 3200 | 3200–5300 | Lognormal | Hall 2009 (base case and range),2 Lively 2019 (range)3 |
Proportion with LRTI | ||||
Age 0–5 mo | 0.65 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Age 6–23 mo | 0.5 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Medically attended outpatient | ||||
Age 0–5 mo | 21 600 | 13 200–21 600 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)34 |
Age 6–11 mo | 24 600 | 17 700–24 600 | Lognormal | |
Age 12–23 mo | 18 440 | 6600–29 620 | Lognormal | Lively 2019 (base case and range),3 Jackson 2021 (range),11 Hall 2009 (range)34 |
Proportion with LRTI | ||||
Age 0–5 mo | 0.65 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Age 6–23 mo | 0.3 | 0.1–1.0 | Beta | Assumption based on Rainisch 202010 |
RSV mortality per hospitalization | ||||
Age 0–5 mo | 0.0010 | 0.0004–0.0020 | Beta | Doucette 2016,12 Hansen 202214 |
Age 6–11 mo | 0.10 | 0.0004–0.0020 | Beta | — |
Age 12–23 mo | 0.003 | 0.0028–0.0034 | Beta | Gupta 201613 |
Intervention efficacy | ||||
Initial efficacy (mo 1–5) against RSV-associated LRTI | 79.0% | 68.5%–86.1% | Beta | Phase 3 trial, phase 2b16 |
Effectiveness mo 6–10 | 25% | 0%–50% | Beta | — |
Effectiveness after 10 mo | 0% | — | — | — |
Adverse events | ||||
Nirsevimab | — | — | — | — |
Probabilities of pediatric adverse events | — | — | — | — |
Systemic reaction | 0.005 | 0.00486–0.00525 | Beta | AstraZeneca ACIP data request |
Probability of outpatient visit given Systemic reaction | 0.1 | — | — | Assumption; Deluca 202322 |
Anaphylaxis | 0 | 0–0.0000010 | — | AstraZeneca ACIP data request |
Injection site reaction | 0.002 | 0.0026–0.0028 | Beta | AstraZeneca ACIP data request |
Probability of outpatient visit given Injection site reaction | 0.1 | — | — | Assumption; Deluca 202322 |
Serious adverse event | 0.000001 | — | Beta | (Guillain-Barre) Prosser 200617 |
Pediatric QALY lost because of adverse events | ||||
Systemic reaction | 0.0056 | 0.00051–0.0061 | Lognormal | Deluca 202322 |
Anaphylaxis | 0.0137 | 0.0135–0.0139 | Lognormal | — |
Serious adverse event | 0.141 | 0.092–0.199 | Lognormal | (Guillain-Barre) Prosser 200617 |
Costs because of adverse events | ||||
Cost of outpatient visit for systemic reaction (non-high–risk) | $313 | $27–$1337 | Lognormal | Marketscan unpublished; Deluca 202322 |
Cost of outpatient visit for injection site reaction | $367.76 | $23.15−$1758 | Lognormal | Marketscan unpublished; Deluca 202322 |
Recipient time for office visit (h) | 2 | 1–3 | Normal | Ray 201535 |
Anaphylaxis medical costs | $7706 | $89–$23 414 | Lognormal | Marketscan unpublished; Deluca 202322 |
Parent time for anaphylaxis (d) | 1 | 1–3 | Normal | Shimabukuo 202136 |
Serious adverse event | $36 163.76 | $10 372.31–$122 145.60 | Lognormal | Prosser 200617 |
Daily productivity for caregivers | 190 | 169.41–211.03 | Lognormal | Grosse 201923 ,b |
Cost inputs | ||||
Palivizumab costs (for high-risk infants not receiving nirsevimab) | ||||
Fraction of children that are high-risk | 1.6% | 0% | 1.6% | Sanofi37 |
Fraction of children at high risk receiving palivizumab | 75% | 0% | 100% | Sobi 202138 |
Palivizumab cost per dose | $1228c | — | — | Shahabi 201839 |
Doses per patient | 4.167d | — | — | Assumption |
RSV costs | ||||
RSV-specific inpatient costs (per inpatient outcome) | ||||
Age 0–11 mo | $11 487 | 4804–86 646 | Lognormal | Bowser 202240 |
Age 12–23 mo | $11 469 | 4804–86 646 | Lognormal | |
D lost productivity | 7.4 | 0–14 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
RSV-specific ED costs (per ED visit) | ||||
Age 0–11 mo | $563 | 544–581 | Lognormal | Bowser 202240 |
Age 12–23 mo | 563 | 544–581 | Lognormal | — |
D lost productivity | 2.5 | 0–5 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
RSV-specific outpatient costs (outpatient visit) | ||||
Age 0–11 mo | $82 | 46–118 | Lognormal | Bowser 20221 |
Age 12–23 mo | $82 | 46–118 | Lognormal | — |
D lost productivity | 2.5 | 0–5 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
Lifetime productivity for those <1 y old | $1 795 936 | 1 346 951–2 244 919 | Lognormal | Grosse 201923 ,e |
Nirsevimab-related costs | ||||
Nirsevimab per dose | $445f | $50–$600 | — | — |
Nirsevimab administration | $22.27 | $16.70–$27.84 | Lognormal | Medicare CPT 96380, administration and counseling |
Proportion of patients requiring specific visit for immunization | 0 | — | — | Assumed part of regularly scheduled office visits |
Discount rate | 0.03 | 0.0–0.07 | — | — |
Quality-of-life lost because of RSV | ||||
Quality-adjusted life d lost from acute RSV | Glaser27 (base case), Regnier28 (lower bound), JIVE COVID-19/RSV utilities (unpublished) | |||
Outpatient | ||||
Child | 3.1 | 1.8–16.6 | Lognormal | |
Caregiver | 1.5 | 0–9.1 | Lognormal | |
ED | ||||
Child | 4.9 | 2.9–16.6g | Lognormal | |
Caregiver | 2.5 | 0–9.1 | Lognormal | |
Hospitalized | ||||
Child | 6.2 | 3.7–26.5 | Lognormal | |
Caregiver | 2.4 | 0–13.6 | Lognormal | |
Discounted QALY lost from death in the | Calculated from y 2020 life tables41 | |||
First year of life | 28.40 | — | — | — |
Second year of life | 28.38 | — | — | — |
Model Input Parameter . | Base case . | Range . | Distribution . | Source . |
---|---|---|---|---|
Category of RSV incidence per 100 000 | ||||
Inpatient | ||||
Age 0 mo | 1760 | 1560–1970 | Lognormal | CDC NVSN4 ,a |
Age 1 mo | 3110 | 2850–3390 | Lognormal | |
Age 2 mo | 2230 | 2030–2450 | Lognormal | |
Age 3 mo | 1560 | 1390–1740 | Lognormal | |
Age 4 mo | 1360 | 1200–1520 | Lognormal | |
Age 5 mo | 1090 | 960–1250 | Lognormal | |
Age 6 mo | 960 | 810–1120 | Lognormal | |
Age 7 mo | 800 | 640–960 | Lognormal | |
Age 8 mo | 730 | 600–880 | Lognormal | |
Age 9 mo | 840 | 680–990 | Lognormal | |
Age 10 mo | 600 | 480–730 | Lognormal | |
Age 11 mo | 600 | 490–730 | Lognormal | |
Age 12 mo | 630 | 500–750 | Lognormal | |
Age 13 mo | 500 | 380–620 | Lognormal | |
Age 14 mo | 580 | 470–700 | Lognormal | |
Age 15 mo | 540 | 430–660 | Lognormal | |
Age 16 mo | 400 | 290–510 | Lognormal | |
Age 17 mo | 370 | 270–460 | Lognormal | |
Age 18 mo | 370 | 260–480 | Lognormal | |
Age 19 mo | 340 | 250–450 | Lognormal | |
Age 20 mo | 280 | 190–370 | Lognormal | |
Age 21 mo | 210 | 140–290 | Lognormal | |
Age 22 mo | 180 | 120–260 | Lognormal | |
Age 23 mo | 290 | 200–380 | Lognormal | |
Proportion with LRTI | ||||
Age 0–5 mo | 1 | 0.5–1.0 | — | Assumption based on Rainisch 202010 |
Age 6–23 mo | 1 | 0.5–1.0 | — | Assumption based on Rainisch 202010 |
ED | ||||
Age 0–5 mo | 7500 | 5500–7500 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)2 |
Age 6–11 mo | 5800 | 5700–5800 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)2 |
Age 12–23 mo | 3200 | 3200–5300 | Lognormal | Hall 2009 (base case and range),2 Lively 2019 (range)3 |
Proportion with LRTI | ||||
Age 0–5 mo | 0.65 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Age 6–23 mo | 0.5 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Medically attended outpatient | ||||
Age 0–5 mo | 21 600 | 13 200–21 600 | Lognormal | Lively 2019 (base case and range),3 Hall 2009 (range)34 |
Age 6–11 mo | 24 600 | 17 700–24 600 | Lognormal | |
Age 12–23 mo | 18 440 | 6600–29 620 | Lognormal | Lively 2019 (base case and range),3 Jackson 2021 (range),11 Hall 2009 (range)34 |
Proportion with LRTI | ||||
Age 0–5 mo | 0.65 | 0.25–1.0 | Beta | Assumption based on Rainisch 202010 |
Age 6–23 mo | 0.3 | 0.1–1.0 | Beta | Assumption based on Rainisch 202010 |
RSV mortality per hospitalization | ||||
Age 0–5 mo | 0.0010 | 0.0004–0.0020 | Beta | Doucette 2016,12 Hansen 202214 |
Age 6–11 mo | 0.10 | 0.0004–0.0020 | Beta | — |
Age 12–23 mo | 0.003 | 0.0028–0.0034 | Beta | Gupta 201613 |
Intervention efficacy | ||||
Initial efficacy (mo 1–5) against RSV-associated LRTI | 79.0% | 68.5%–86.1% | Beta | Phase 3 trial, phase 2b16 |
Effectiveness mo 6–10 | 25% | 0%–50% | Beta | — |
Effectiveness after 10 mo | 0% | — | — | — |
Adverse events | ||||
Nirsevimab | — | — | — | — |
Probabilities of pediatric adverse events | — | — | — | — |
Systemic reaction | 0.005 | 0.00486–0.00525 | Beta | AstraZeneca ACIP data request |
Probability of outpatient visit given Systemic reaction | 0.1 | — | — | Assumption; Deluca 202322 |
Anaphylaxis | 0 | 0–0.0000010 | — | AstraZeneca ACIP data request |
Injection site reaction | 0.002 | 0.0026–0.0028 | Beta | AstraZeneca ACIP data request |
Probability of outpatient visit given Injection site reaction | 0.1 | — | — | Assumption; Deluca 202322 |
Serious adverse event | 0.000001 | — | Beta | (Guillain-Barre) Prosser 200617 |
Pediatric QALY lost because of adverse events | ||||
Systemic reaction | 0.0056 | 0.00051–0.0061 | Lognormal | Deluca 202322 |
Anaphylaxis | 0.0137 | 0.0135–0.0139 | Lognormal | — |
Serious adverse event | 0.141 | 0.092–0.199 | Lognormal | (Guillain-Barre) Prosser 200617 |
Costs because of adverse events | ||||
Cost of outpatient visit for systemic reaction (non-high–risk) | $313 | $27–$1337 | Lognormal | Marketscan unpublished; Deluca 202322 |
Cost of outpatient visit for injection site reaction | $367.76 | $23.15−$1758 | Lognormal | Marketscan unpublished; Deluca 202322 |
Recipient time for office visit (h) | 2 | 1–3 | Normal | Ray 201535 |
Anaphylaxis medical costs | $7706 | $89–$23 414 | Lognormal | Marketscan unpublished; Deluca 202322 |
Parent time for anaphylaxis (d) | 1 | 1–3 | Normal | Shimabukuo 202136 |
Serious adverse event | $36 163.76 | $10 372.31–$122 145.60 | Lognormal | Prosser 200617 |
Daily productivity for caregivers | 190 | 169.41–211.03 | Lognormal | Grosse 201923 ,b |
Cost inputs | ||||
Palivizumab costs (for high-risk infants not receiving nirsevimab) | ||||
Fraction of children that are high-risk | 1.6% | 0% | 1.6% | Sanofi37 |
Fraction of children at high risk receiving palivizumab | 75% | 0% | 100% | Sobi 202138 |
Palivizumab cost per dose | $1228c | — | — | Shahabi 201839 |
Doses per patient | 4.167d | — | — | Assumption |
RSV costs | ||||
RSV-specific inpatient costs (per inpatient outcome) | ||||
Age 0–11 mo | $11 487 | 4804–86 646 | Lognormal | Bowser 202240 |
Age 12–23 mo | $11 469 | 4804–86 646 | Lognormal | |
D lost productivity | 7.4 | 0–14 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
RSV-specific ED costs (per ED visit) | ||||
Age 0–11 mo | $563 | 544–581 | Lognormal | Bowser 202240 |
Age 12–23 mo | 563 | 544–581 | Lognormal | — |
D lost productivity | 2.5 | 0–5 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
RSV-specific outpatient costs (outpatient visit) | ||||
Age 0–11 mo | $82 | 46–118 | Lognormal | Bowser 20221 |
Age 12–23 mo | $82 | 46–118 | Lognormal | — |
D lost productivity | 2.5 | 0–5 | Lognormal | Fragaszy 2018,24 Petrie 2016,25 Van Wormer 201726 |
Lifetime productivity for those <1 y old | $1 795 936 | 1 346 951–2 244 919 | Lognormal | Grosse 201923 ,e |
Nirsevimab-related costs | ||||
Nirsevimab per dose | $445f | $50–$600 | — | — |
Nirsevimab administration | $22.27 | $16.70–$27.84 | Lognormal | Medicare CPT 96380, administration and counseling |
Proportion of patients requiring specific visit for immunization | 0 | — | — | Assumed part of regularly scheduled office visits |
Discount rate | 0.03 | 0.0–0.07 | — | — |
Quality-of-life lost because of RSV | ||||
Quality-adjusted life d lost from acute RSV | Glaser27 (base case), Regnier28 (lower bound), JIVE COVID-19/RSV utilities (unpublished) | |||
Outpatient | ||||
Child | 3.1 | 1.8–16.6 | Lognormal | |
Caregiver | 1.5 | 0–9.1 | Lognormal | |
ED | ||||
Child | 4.9 | 2.9–16.6g | Lognormal | |
Caregiver | 2.5 | 0–9.1 | Lognormal | |
Hospitalized | ||||
Child | 6.2 | 3.7–26.5 | Lognormal | |
Caregiver | 2.4 | 0–13.6 | Lognormal | |
Discounted QALY lost from death in the | Calculated from y 2020 life tables41 | |||
First year of life | 28.40 | — | — | — |
Second year of life | 28.38 | — | — | — |
Distribution is used for the probabilistic sensitivity analysis. The mean is the same as the base case value with an SD of one-quarter of the range. Efficacy estimates are based on pooled estimates from the phase 2b and phase 3 trials. In the phase 2b trial, all infants received 50 mg of nirsevimab. Because nirsevimab is approved for a dose of 50 mg for infant weighing <5 kg and for a dose of 100 mg for infants weighing ≥5 kg or more, pooled efficacy estimated excluded infants who weighed ≥5 kg in the phase 2b trial. The phase 3 trial used the approved dosing regimen. A QALY is 1 365th of a QALY (QALY efficacy is for hospitalization, ED, and outpatient outcomes, but it is assumed to be effective against LRTI only in the base case; we assume 0% efficacy against upper-respiratory outcomes). CDC, Centers for Disease Control and Prevention; COVID-19, coronavirus disease 2019; CPT, Current Procedural Terminology; JIVE, Joint Initiative in Vaccine Economics; NVSN, New Vaccine Surveillance Network.
CDC New Vaccine Surveillance Network hospitalization rates for children <2 years of age from December 2016 to September 2020.
Daily productivity rate calculated by dividing mean annual total productivity (both market and nonmarket) for ages 15 to 99 by 365 days and inflated from 2016 to 2022 value using the Federal Reserve Gross Domestic Product implicit price deflator.
Palivizumab cost per dose is the drug costs. We do not include costs (or quality-of-life losses) associated with adverse events from palivizumab administration.
Based on an assumption 66% receive 5 doses and 8.3% each receive 4, 3, 2, and 1 doses.
Lifetime productivity taken from Table 2 of Grosse for age 0 lifetime total productivity (both market and nonmarket) assuming 1% annual productivity growth and a 3% discount rate and inflated from 2016 to 2022 dollars using the Federal Reserve GDP implicit price deflator.
The cost of nirsevimab was based on an anticipated commercial price of $495 and a Vaccines for Children price of $395. We assumed 50% of doses would be purchased through Vaccines for Children to calculate an average of $445 per dose as our base case cost.
Upper bound for sensitivity analysis utilities was not directly measured for ED in the JIVE/coronavirus disease 2019 RSV utilities study, but outpatient values were used.
Efficacy and Safety Inputs
Nirsevimab efficacy was estimated on the basis of pooled results from the phase 2b and phrase 3 clinical trials (Table 1).17 After administration, we assumed that efficacy follows a sigmoid function over the first 150 days, declining to an efficacy of 25% by day 150 and then reaching 0% efficacy at month 10 (Supplemental Fig 5). The average efficacy in the first 150 days matches the reported efficacy against RSV-associated MA-LRTI. Because the trial results measured efficacy against RSV MA-LRTI and efficacy against URTI is unknown, we assumed that nirsevimab only protected against RSV MA-LRTI and offered no protection against RSV URTI. In a sensitivity analysis, we assumed flat efficacy during days 0 through 150, then immediately declining to 0% efficacy. Rates of adverse events including systemic reaction and injection site reaction came from clinical trials (Table 1). Severe adverse events (eg, anaphylaxis or Guillain-Barre) were not observed in trials and, in general, rare events may not be observed in clinical trials because of the limited number of enrolled participants. However, to account for unobserved risks in larger populations, we assumed a 1 in 1 million rate of hypothetical serious adverse events.18
Costs Inputs
The model included costs associated with nirsevimab and RSV MA-LRTI outcomes. The cost of nirsevimab was based on an anticipated commercial price of $495 and a Vaccines for Children price of $395.19 We assumed 50% of doses would be purchased through Vaccines for Children,20–22 so we used a weighted average of $445 per dose as our base case cost (Table 1). Because the dose was anticipated to be given at birth or a routine outpatient visit (well-child checkup), we did not include costs of an extra health care visit to administer nirsevimab, but we included $22.27 for counseling and administration. We based adverse event costs on a recent analysis of MarketScan data23 and included productivity losses from workdays lost by caregivers for these adverse events (Table 1).24 The comparator arm of no nirsevimab assumes a standard of care that includes palivizumab use among eligible high-risk infants (Table 1). In the nirsevimab strategy, uptake of nirsevimab was assumed to be the same in all infants, including those that are palivizumab eligible, so eligible high-risk infants who do not receive nirsevimab may still receive palivizumab.
Costs for RSV came from a systematic literature review of RSV costs in infants (Table 1).1 To account for total societal costs, we also included days of work lost for caregivers and the dollar value of lifetime medical costs and productivity losses in infants associated with RSV-related premature deaths using the human capital approach (Table 1 and Supplemental Table 5).25–27
Quality of Life Inputs
The health-related quality of life associated with RSV outcomes of hospitalization, ED visits, and outpatient visits in infants is particularly challenging to evaluate because infants cannot be queried directly, and there may be spillover effects on families. So, in addition to quality-adjusted life years (QALYs) lost to the child, we also incorporated QALYs lost from caregivers. We used a systematic review of literature on child and caregiver quality of life28 for our base case QALYs losses. However, for the lower-bound values, we resorted to a study by Regnier29 and to an unpublished preference elucidation survey of parents of RSV patients for our upper-bound values (Table 1).
Analysis Plan
Health and Economic Outcomes
The model simulated RSV MA-LRTI–associated outpatient visits, ED visits, hospitalizations, and deaths to calculate total medical and productivity costs and QALYs lost. The main summary measure is the incremental cost-effectiveness ratio (ICER) in terms of dollars per QALYs gained. This metric tells us how much money is spent per year of life in perfect health (QALY) gained with the use of nirsevimab. Assuming spending less on health is better, a lower ICER is preferred. We also estimated the number needed to immunize to prevent an undesirable RSV disease outcome (eg, number of infants needed to be immunized with nirsevimab to prevent an RSV-associated hospitalization or death). Other intermediate outcomes included numbers of outpatient visits, ED visits, hospitalizations, medical and indirect costs, and total cost per outcome averted. Costs are reported in 2022 US dollars for a 1-year time frame, and future costs and health outcomes are discounted at 3% annually. We used Microsoft Excel 365 for all calculations.
Primary Analysis: Infants Born During or Entering Their First RSV Season
The primary analysis estimated the ICER of nirsevimab compared with no nirsevimab in the first RSV season.
Secondary Analysis: High-Risk Children Entering Their Second RSV Season
The secondary analysis evaluated the ICER of nirsevimab compared with no nirsevimab in the month of October before the second RSV season for children 8 to 19 months of age for a variety of risk levels. Few data are available on the incidence of RSV MA-LRTI and RSV hospitalization by risk condition. For this analysis, we created theoretical risk groups that had an RSV-associated hospitalization incidence of 2, 3, 6, or 10 times the incidence in the general population. We conservatively assumed that the incidence of outpatient visits, ED visits, death per hospitalization, and cost of health care visits were the same as those of the general population. We assumed children currently eligible for palivizumab in their second season would already receive nirsevimab, given the higher efficacy, lower cost, and implementation profile of nirsevimab (single dose versus monthly dosing). Therefore, we focused the analysis for the second season on children not already eligible for palivizumab, and thus savings from palivizumab were not included in this analysis.30–32 We assumed the cost of nirsevimab is doubled because 2 100 mg doses of nirsevimab would be required in this group.
Sensitivity and Scenario Analyses
We conducted 1-way, 2-way, and probabilistic sensitivity analyses. Parameter ranges for sensitivity analysis can be found in Table 1, as well as Supplemental Fig 4. Probabilistic sensitivity analysis assigned distributions to input parameters (see Table 1) and using 1000 Monte Carlo simulation iterations, calculated overall uncertainty using cost-effectiveness acceptability curves. The results of these simulations are also used to create 95% credible intervals for outcome estimates. We also explored scenarios where there were no savings from palivizumab (a policy for a population not currently eligible to receive palivizumab and where there was lower risk of mortality), and where nirsevimab might have an equivalent prevention efficacy on URTI as on LRTI. Finally, we explored an additional tertiary analysis where we assume the pregnant mother had received the US Food and Drug Administration-approved RSVpreF vaccine more than 14 days before delivery (inputs in Supplemental Table 6).10 In that scenario, because there are no data available on the efficacy of nirsevimab when given to infants born to vaccinated mothers, we assumed the efficacy of adding nirsevimab on top of RSVpreF vaccine protection would be the higher efficacy of either nirsevimab or RSVpreF (Supplemental Fig 6).
Results
Outcomes, Outcomes Averted, and Number Needed to Immunize
Without nirsevimab, our model projects RSV-associated illness among the annual US birth cohort would result in 47 758 hospitalizations, with 48 deaths occurring among inpatients (Table 2). If nirsevimab was given in the first RSV season to half of an annual US birth cohort, our model projects that 30% of hospitalization visits and deaths would be averted (Table 2). It would take 128 immunized with nirsevimab to prevent a hospitalization.
Cost-Effectiveness Outcomes for Use of Nirsevimab
. | RSV Outcomes . | ||||||
---|---|---|---|---|---|---|---|
LRTI or Total . | No. Outpatient Visits . | No. ED visits . | No. Inpatient Visits . | No. Deaths . | Costs (Millions) . | QALYs Lost . | |
Natural history | LRTI | 392 446 (143 969–641 886) | 142 449 (71 910–198 938) | 47 758 (41 803–53 998) | 48 (18–96) | — | — |
Total | 846 451 (570 447–881 128) | 243 675 (205 206–249 996) | 47 758 (41 803–53 998) | 48 (18–96) | 1651 (857–3857) | 18 151 (6844–42 113) | |
Nirsevimaba | LRTI | 285 193 (97 840–482 578) | 104 245 (52 497–146 998) | 33,417 (27 666–39 451) | 33 (15–81) | — | — |
Total | 739 199 (506 623–800 399) | 205 471 (169 765–224 312) | 33 417 (27 666–39 451) | 33 (12–67) | 2085 (1447–3607) | 15 324 (5569–36 362) | |
Difference | LRTI | 107 253 (−178 103 to −40 827) | 38 204 (−59 478 to −17 207) | 14 341 (−18 999 to −10 672) | 14 (−30 to −5) | — | — |
Total | 107 253 (−178 103 to −40 827) | 38 204 (−59 478 to −17 207) | 14 341 (−18 999 to −10 672) | 14 (−30 to −5) | 434 (−292 to –623) | −2827 (−6578 to −1 075) | |
Cost per outcome averted | Total | 4047 (−2756–12 922) | 11 361 (−7457 to 30 127) | 30 264 (−19 111–52 294) | 30 263 954 (−19 309 to 397 815 665) | N/A | 153 517 (−106 281 to 490 747) |
. | RSV Outcomes . | ||||||
---|---|---|---|---|---|---|---|
LRTI or Total . | No. Outpatient Visits . | No. ED visits . | No. Inpatient Visits . | No. Deaths . | Costs (Millions) . | QALYs Lost . | |
Natural history | LRTI | 392 446 (143 969–641 886) | 142 449 (71 910–198 938) | 47 758 (41 803–53 998) | 48 (18–96) | — | — |
Total | 846 451 (570 447–881 128) | 243 675 (205 206–249 996) | 47 758 (41 803–53 998) | 48 (18–96) | 1651 (857–3857) | 18 151 (6844–42 113) | |
Nirsevimaba | LRTI | 285 193 (97 840–482 578) | 104 245 (52 497–146 998) | 33,417 (27 666–39 451) | 33 (15–81) | — | — |
Total | 739 199 (506 623–800 399) | 205 471 (169 765–224 312) | 33 417 (27 666–39 451) | 33 (12–67) | 2085 (1447–3607) | 15 324 (5569–36 362) | |
Difference | LRTI | 107 253 (−178 103 to −40 827) | 38 204 (−59 478 to −17 207) | 14 341 (−18 999 to −10 672) | 14 (−30 to −5) | — | — |
Total | 107 253 (−178 103 to −40 827) | 38 204 (−59 478 to −17 207) | 14 341 (−18 999 to −10 672) | 14 (−30 to −5) | 434 (−292 to –623) | −2827 (−6578 to −1 075) | |
Cost per outcome averted | Total | 4047 (−2756–12 922) | 11 361 (−7457 to 30 127) | 30 264 (−19 111–52 294) | 30 263 954 (−19 309 to 397 815 665) | N/A | 153 517 (−106 281 to 490 747) |
Negative numbers in the difference row indicate gains or events saved with the nirsevimab arm as compared with natural history. Positive numbers mean worse outcomes like increased costs. LRTI, Lower Respiratory Tract Infection; N/A, not applicable; No., number; RSV, Respiratory Syncytial Virus; QALY, Quality-Adjusted Life-Year. —, costs and QALYs lost are only reported in totals.
Nirsevimab given to 50% of US infants aged <8 months entering their first RSV season.
Cost-Effectiveness
Without nirsevimab, $225 million would be spent on palivizumab for roughly 1.2% of the birth cohort, $755 million on RSV medical care, and $671 million on productivity losses, for $1651 million in total costs (Table 3). A combination of nirsevimab immunization for 50% of the birth cohort and palivizumab for 0.6% (half of the 1.2% who did not receive nirsevimab) would lead to $969 million in intervention costs (330% higher), $560 million in RSV medical care (26% lower), and $556 million in productivity costs (17% lower) for $2085 million in total costs (26% higher) (Table 3). Although costs are higher with nirsevimab, the model projects 2827 more net QALYs gained (Table 4), leading to a societal ICER of $153 517 per QALY gained (Table 1). Costs per event averted are in Table 1.
Resulting Costs (in Millions of 2022 US Dollars) in US Infants Aged <8 Months Entering Their First RSV Season, Without and With Nirsevimab Among 50% of US Birth Cohort
. | Medical Costs (Millions of 2022 US Dollars) . | Productivity Costs (Millions of 2022 US Dollars) . | . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Intervention Costs . | Outpatient Costs . | ED Costs . | Inpatient Costs . | Total RSV Medical Costs . | Total Health System Costs . | Outpatient Costs . | ED Costs . | Inpatient Costs . | Death Costs . | Total Productivity Costs . | Total Costs . |
Natural history | 225 (103–398) | 69 (38–102) | 137 (116–142) | 549 (25–2716) | 755 (217–2932) | 980 (387–3166) | 402 (122–876) | 116 (39–260) | 67 (24–146) | 86 (31–175) | 671 (338–1164) | 1651 (857–3857) |
Nirsevimaba | 969 (868–1,023) | 61 (33–89) | 116 (95–127) | 384 (17–1908) | 560 (179–2078) | 1529 (1087–3024) | 351 (105–762) | 98 (32–219) | 47 (17–101) | 60 (21–122) | 556 (270–987) | 2085 (1447–3607) |
Difference | 744 (619–777) | −9 (−17 to −3) | −22 (−34 to −10) | −165 (−838 to −8) | −195 (−870 to −33) | 549 (−158 to 711) | −51 (−131 to −11) | −18 (−46 to −5) | −20 (−46 to −7) | −26 (−56 to −9) | −115 (−208 to −57) | 434 (−292 to 623) |
. | Medical Costs (Millions of 2022 US Dollars) . | Productivity Costs (Millions of 2022 US Dollars) . | . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Intervention Costs . | Outpatient Costs . | ED Costs . | Inpatient Costs . | Total RSV Medical Costs . | Total Health System Costs . | Outpatient Costs . | ED Costs . | Inpatient Costs . | Death Costs . | Total Productivity Costs . | Total Costs . |
Natural history | 225 (103–398) | 69 (38–102) | 137 (116–142) | 549 (25–2716) | 755 (217–2932) | 980 (387–3166) | 402 (122–876) | 116 (39–260) | 67 (24–146) | 86 (31–175) | 671 (338–1164) | 1651 (857–3857) |
Nirsevimaba | 969 (868–1,023) | 61 (33–89) | 116 (95–127) | 384 (17–1908) | 560 (179–2078) | 1529 (1087–3024) | 351 (105–762) | 98 (32–219) | 47 (17–101) | 60 (21–122) | 556 (270–987) | 2085 (1447–3607) |
Difference | 744 (619–777) | −9 (−17 to −3) | −22 (−34 to −10) | −165 (−838 to −8) | −195 (−870 to −33) | 549 (−158 to 711) | −51 (−131 to −11) | −18 (−46 to −5) | −20 (−46 to −7) | −26 (−56 to −9) | −115 (−208 to −57) | 434 (−292 to 623) |
Negative numbers in the difference row indicate savings with the nirsevimab arm as compared with natural history. The natural history comparator policy intervention costs include palivizumab use in this cost category. Because we assume 50% of infants get nirsevimab in the nirsevimab strategy, the other 50% might (if they are high risk) get palivizumab. Therefore, the nirsevimab intervention number includes both nirsevimab and palivizumab costs.
Nirsevimab given to 50% of US infants aged <8 months entering their first RSV season.
Resulting Quality Adjusted Life Years (QALYs) Lost From RSV in US Infants Aged <8 Months Entering Their First RSV Season and Their Caregivers, Without and With Nirsevimab Among 50% of US Birth Cohort
. | Adverse Events . | Outpatient . | ED . | Inpatient . | Deaths . | Total . | Grand Total . | |||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Child . | Caregiver . | Child . | Caregiver . | Child . | Caregiver . | Child . | Child . | Caregiver . | ||||
Natural history | 0 (0–0) | 7153 (628–26 624) | 3580 (213–17 252) | 3290 (760–9012) | 1645 (238–5543) | 807 (127–2861) | 320 (24–1352) | 1356 (509–2717) | 12 606 (4130–32 528) | 5545 (1053–19 610) | 18 151 (6844–42 113) | |
Nirsevimab to 50% of the birth cohort | 52 (47–58) | 6246 (548–23 313) | 3127 (189–15 166) | 2774 (636–7519) | 1387 (198–4637) | 565 (87–1993) | 224 (17–949) | 949 (350–1900) | 10 586 (3317–27 793) | 4738 (870–16 892) | 15 324 (5569–36 362) | |
QALYs gained | −52 (−58 to −47) | 906 (65–3663) | 454 (23–2323) | 516 (96–1555) | 258 (31–946) | 242 (38–885) | 96 (7–424) | 407 (148–862) | 2020 (710–5012) | 808 (156–2811) | 2827 (1075–6578) |
. | Adverse Events . | Outpatient . | ED . | Inpatient . | Deaths . | Total . | Grand Total . | |||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Child . | Caregiver . | Child . | Caregiver . | Child . | Caregiver . | Child . | Child . | Caregiver . | ||||
Natural history | 0 (0–0) | 7153 (628–26 624) | 3580 (213–17 252) | 3290 (760–9012) | 1645 (238–5543) | 807 (127–2861) | 320 (24–1352) | 1356 (509–2717) | 12 606 (4130–32 528) | 5545 (1053–19 610) | 18 151 (6844–42 113) | |
Nirsevimab to 50% of the birth cohort | 52 (47–58) | 6246 (548–23 313) | 3127 (189–15 166) | 2774 (636–7519) | 1387 (198–4637) | 565 (87–1993) | 224 (17–949) | 949 (350–1900) | 10 586 (3317–27 793) | 4738 (870–16 892) | 15 324 (5569–36 362) | |
QALYs gained | −52 (−58 to −47) | 906 (65–3663) | 454 (23–2323) | 516 (96–1555) | 258 (31–946) | 242 (38–885) | 96 (7–424) | 407 (148–862) | 2020 (710–5012) | 808 (156–2811) | 2827 (1075–6578) |
Negative numbers in the QALYs gained row indicate losses with the Nirsevimab arm as compared with natural history.
In our secondary analysis focused on the second RSV season, the ICER depends greatly on the risk of severe outcomes. For an average risk child, receiving nirsevimab costs >$1.6 million per QALY gained; however, for young children with a 10-fold risk of hospitalization, the cost would be $308 468 per QALY gained (Supplemental Figs 7 and 8).
Sensitivity and Scenario Analyses
The results are sensitive to several parameters (Fig 2). If the cost of nirsevimab dropped to $50 per dose, nirsevimab was cost-saving, but if it rose to $600 per dose, the cost rose to $253 964 per QALY gained (Fig 3). The range of costs per hospitalization were wide (ie, a base case of $11 487 but ranging from $4804 to $86 646). Using the highest cost assumption, use of nirsevimab in infants aged <8 months became cost-saving. The quality-of-life associated with RSV outcomes also had a big impact. If RSV has a larger burden for children and their caregivers (ie, the highest QALY losses per RSV-associated hospitalization, ED, and outpatient), then the cost dropped to $35 659 per QALY gained. However, if the impact of RSV on quality of life is low, then the cost increased to $323 788 per QALY gained. Efficacy of nirsevimab was important: if it was high, the ICER was $98 183 per QALY gained, and if low, it was $243 556 per QALY gained. If nirsevimab were just as effective at preventing URTI as it has been shown to prevent MA-LRTI, then the estimated societal cost would be $78 420 per QALY gained (Fig 3). If efficacy was a flat 79% efficacy through day 150 and then immediately dropping to 0%, the ICER would be $152 651 per QALY gained. Finally, the use of palivizumab had an impact on the ICER because nirsevimab was assumed to reduce palivizumab use. If none of the population under consideration needed palivizumab, the ICER rose to $193 310 per QALY gained (Fig 2).
One-way sensitivity analysis. Note: The ICER measures the cost per QALY saved with nirsevimab use in US infants aged <8 months. The colored bars show how the ICER can change as the parameter assumptions change. This figure ranks parameters by their influence on the base case ICER. For example, if the inpatient costs associated with RSV are low (blue bar), then the ICER rises to $181 908 per QALY gained. If the initial efficacy high (orange bar), then the ICER drops such that using nirsevimab is cost-saving. Bars that go to 0 mean nirsevimab is cost-saving. “Fraction receiving palivizumab natural history” is the fraction of infants receiving palivizumab in the natural history arm. If many children are receiving expensive palivizumab in the natural history arm, there is a greater potential to avoid that expense if nirsevimab is used instead.
One-way sensitivity analysis. Note: The ICER measures the cost per QALY saved with nirsevimab use in US infants aged <8 months. The colored bars show how the ICER can change as the parameter assumptions change. This figure ranks parameters by their influence on the base case ICER. For example, if the inpatient costs associated with RSV are low (blue bar), then the ICER rises to $181 908 per QALY gained. If the initial efficacy high (orange bar), then the ICER drops such that using nirsevimab is cost-saving. Bars that go to 0 mean nirsevimab is cost-saving. “Fraction receiving palivizumab natural history” is the fraction of infants receiving palivizumab in the natural history arm. If many children are receiving expensive palivizumab in the natural history arm, there is a greater potential to avoid that expense if nirsevimab is used instead.
Impact of varying nirsevimab cost per dose on the ICER for preventing RSV LRTI (base case) or when assuming equal efficacy also preventing RSV URTI (scenario). The dots represent the base case cost per dose of $445.
Impact of varying nirsevimab cost per dose on the ICER for preventing RSV LRTI (base case) or when assuming equal efficacy also preventing RSV URTI (scenario). The dots represent the base case cost per dose of $445.
We examined several scenarios. If we exclude productivity costs to evaluate health system costs alone, the ICER increases to $194 198 per QALY gained (Supplemental Table 7). In 1 scenario, we examined the combined effects of RSV mortality and savings because of reduced palivizumab use (Supplemental Table 8). In the first part of this scenario, we considered infants at low risk of dying. If the population were not high risk (ie, with a 0.04% risk of death, if hospitalized), then the societal cost increased to $174 026 per QALY gained. In the second part of this scenario, we examined palivizumab use. If administering nirsevimab did not replace palivizumab (eg, in a population not initially eligible for palivizumab), then the cost rose to $193 310 per QALY gained. Finally, in the combined part of this scenario, if the population had low risk of mortality and would have never received palivizumab, then the cost for receiving nirsevimab was $217 584 per QALY gained. Additional scenarios related to the timing of administration of nirsevimab are shown in Supplemental Figs 9 and 10.
In the probabilistic sensitivity analysis of the primary analysis, we found 20% of simulations had an incremental cost of <$100 000 per QALY gained, 54% of simulations had an incremental cost of <$200 000 per QALY gained, and 97% had an incremental cost of <$500 000 per QALY gained (Supplemental Fig 11).
Finally, we examined a scenario where we assumed the pregnant mother had received an RSVpreF vaccine >14 days before delivery. In that scenario, the societal cost increased to $435 114 per QALY gained because the marginal benefit of adding nirsevimab protection above the existing RSVpreF protection was smaller (Supplemental Fig 12). However, the value of adding nirsevimab on top of RSVpreF protection varied by the month of birth of the child and the risk level of the child (Supplemental Fig 13).
Discussion
Our analysis shows administering nirsevimab to infants <8 months of age before entering their first RSV season in the United States is expected to decrease RSV LRTI-associated medical events (ie, outpatient and ED patient visits, hospitalizations, and deaths) and could potentially be a cost-effective way to reduce RSV burden in infants in their first RSV season with a societal cost of $153 517 per QALY gained. The results of this analysis were incorporated into the ACIP’s decisions for recommendations for infants and young children in their first RSV season.8 Administering nirsevimab to all young children in their second season is unlikely to be cost-effective, but it may be cost-effective for certain young children at a significantly higher risk of severe outcomes from RSV (eg, chronic lung disease or severe immunocompromise). On the basis of these results and other data, the ACIP recommended nirsevimab for the second RSV season only in children who the American Academy of Pediatrics deems eligible for palivizumab in their second year of life.8 ACIP also recommended nirsevimab for American Indian and Alaska Native children entering their second RSV season because of studies showing an increased risk of RSV hospitalization in this population.6
There are no other published cost-effectiveness studies on nirsevimab in US infants. However, another unpublished and manufacturer-sponsored cost-effectiveness analysis was also discussed and summarized for the ACIP’s consideration in February 2023.33 That analysis had similarities in modeling approach with our model, but came to the conclusion that nirsevimab had a societal cost of $70 430 per QALY gained. That analysis incorporated assumptions more favorable for nirsevimab cost-effectiveness compared with the analysis in our study. Higher incidence of hospitalization, much higher costs of hospitalization, and higher costs associated with outpatient visits were among the more influential assumptions.
The cost-effectiveness of administering nirsevimab to infants is sensitive to a variety of factors. The cost per dose of nirsevimab could be highly influential as to whether nirsevimab is cost-effective or not. Uncertainty in RSV-associated health care costs and quality-of-life remain sizable and highly correlated with nirsevimab cost-effectiveness. If RSV hospitalizations were more expensive, or if quality-of-life losses from RSV were higher, then nirsevimab may be more cost-effective.
Our results on the cost-effectiveness of administering nirsevimab to infants of vaccinated mothers should be interpreted with caution; the added protection of adding nirsevimab is unknown. Our base case ICER would apply to infants whose gestational parent was not vaccinated with RSVpreF within 14 days of delivery, and are given nirsevimab. However, our preliminary scenario analysis suggests that administering nirsevimab to infants whose mothers received RSVpreF vaccine ≥14 days before giving birth may be much less cost-effective than when administered to unvaccinated mothers, but potentially could be cost-effective if administered to the highest-risk populations, or if nirsevimab provides increased or longer protection against RSV-associated MA-LRTI. ACIP recommended that nirsevimab may be considered for infants born to vaccinated mothers if the health care provider assesses that additional protection is warranted because the infant is at substantial increased risk for severe RSV disease.
A natural question is how do nirsevimab and RSVpreF immunization compare when used alone. In our analysis of maternal RSVpreF immunization, we found ICERs of $163 513 per QALY saved when administered September through January or $396 280 when administered yearround, suggesting nirsevimab may be more cost-effective compared with maternal RSVpreF immunization alone;10 however, differences in the trials may make the products difficult to directly compare.
There are several limitations to our analysis. Although in some scenarios we evaluate varying levels of hospitalization risks, our model structure does not include specific clinically-defined risk groups because of lack of available data. We also do not include any impact on reducing person-to-person transmission of RSV. If nirsevimab reduced RSV transmission (to other infants or to the elderly), nirsevimab might be more cost-effective. Although a theoretical model of RSV transmission suggests that vaccination of infants may reduce transmission to other groups,34 that study is based on theoretical assumptions seldom supported by epidemiologic studies that trace RSV transmission, and we lack clinical evidence that nirsevimab (or future RSV vaccine products for these age groups) reduces transmission. In analyzing the cost-effectiveness of immunization in the second season for children at high risk, our assumption that risks of outpatient visits, ED visits, or deaths given hospitalization did not increase may underestimate costs and health outcomes for individuals with high-risk conditions, thus underestimating the value of nirsevimab immunization. Additional research on RSV disease transmission along with its medical costs, quality-of-life impact, as well as real-world performance of nirsevimab, may be helpful to better understand the health benefits and cost-effectiveness of nirsevimab.
Conclusions
Nirsevimab to prevent RSV MA-LRTI in infants and young children is likely to sizably decrease RSV disease burden. Although subject to key factors, nirsevimab could also be societally cost-effective for all infants entering the first RSV season, and in the second season for certain young children facing a greater risk of severe disease.
Dr Hutton conceptualized and designed the study, conducted the analyses, and drafted the initial manuscript; Drs Prosser, Ortega-Sanchez, Leidner, and Pike conceptualized and designed the study; Ms Rose and Ms Mercon collected data and drafted the initial manuscript; Drs McMorrow, Fleming-Dutra, Prill, and Jones conceptualized and designed the study and collected data; and all authors critically reviewed and revised the manuscript for important intellectual content, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
COMPANION PAPERS: Companions to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2024-066481 and www.pediatrics.org/cgi/doi/10.1542/peds.2024-068261.
FUNDING: Supported by Centers for Disease Control and Prevention contract 75D30122P15319 respiratory syncytial virus modeling, cost assessments, and health utility analysis. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest relevant to this article to disclose. The opinions expressed by the authors do not necessarily reflect the opinions of the Centers for Disease Control and Prevention.
Comments