(23) Modelling the cost-effectiveness of catch-up 'MenB' (Bexsero) vaccination in England

Home

Questions

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

Papers

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

PMCID: 5418563 (link)

Year: 2017

Reviewer Paper ID: 23

Project Paper ID: 89

Q1 - Title
(show question description)

Question description: Does the title clearly identify the study as an economic evaluation and specify the interventions being compared?

Explanation: The title of the manuscript does not specify the interventions being compared as it only mentions a single intervention, the 'catch-up MenB (Bexsero) vaccination.' It also does not explicitly describe the study as an economic evaluation, although it implies cost considerations with the term 'cost-effectiveness.'

Quotes:

  • Modelling the cost-effectiveness of catch-up 'MenB' (Bexsero) vaccination in England

Q2 - Abstract
(show question description)

Question description: Does the abstract provide a structured summary that includes the context, key methods, results, and alternative analyses?

Explanation: The abstract does not provide a structured summary that includes context, key methods, results, and alternative analyses in the style of structured abstracts. Instead, it is primarily a collection of statements focusing on conclusions and high-level results without explicit sections for context, methods, and alternative analyses.

Quotes:

  • We modelled the cost-effectiveness catch-up 'MenB' vaccination in England.
  • Catch-up vaccination for 1 year old children could be cost-effective at <=$8 per dose.
  • Additionally vaccinating 2 year olds was less cost-effective.
  • With conservative vaccine assumptions vaccinating 2 year olds was not-cost-effective.
  • It was not cost-effective to extend vaccination further to 3-4 year olds.
  • We assessed the cost-effectiveness of offering catch-up vaccination with Bexsero against meningococcal disease...
  • Offering catch-up vaccination to increasingly older children is less economically attractive because of declining disease burden.

Q3 - Background and objectives
(show question description)

Question description: Does the introduction provide the context for the study, the study question, and its practical relevance for decision-making in policy or practice?

Explanation: The introduction provides the context by detailing the UK policy decision to offer the MenB vaccine to infants and the public's demand for wider coverage. It also addresses the study question by outlining the uncertainty regarding the cost-effectiveness of catch-up vaccination in older children, setting up the investigation into optimal catch-up strategies based on disease burden.

Quotes:

  • 'In September 2015 the UK became the first country in the world to routinely offer to infants a vaccine (Bexsero) against MenB disease at 2, 4 and 12 months of age.'
  • 'The decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive disease.'
  • 'The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options...'

Q4 - Health economic analysis plan
(show question description)

Question description: Was a health economic analysis plan developed, and if so, where is it available?

Explanation: The manuscript does not mention the development of a specific health economic analysis plan, nor does it reference the availability of such a plan. Instead, it discusses the use of a previously developed model that was 'independently reviewed' as the basis for the current economic analysis.

Quotes:

  • A strength of this work is that it builds upon a previously independently reviewed model which was used to inform the infant recommendation for the use of Bexsero in the UK.

Q5 - Study population
(show question description)

Question description: Are the characteristics of the study population (e.g., age range, demographics, socioeconomic, or clinical characteristics) described?

Explanation: The manuscript does not provide detailed descriptions of the study population's characteristics such as age range beyond certain age groups, demographics, or socioeconomic, or clinical characteristics. It focuses mainly on the economic modeling of different vaccination strategies.

Quotes:

  • The article mentions the focus on age groups: '...focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.'
  • The incidence is provided without further demographic details: '...driven by the fact that disease incidence in 2 year olds is lower than in 1 year old children; ... in the absence of vaccination the model assumes 359 cases in infants, 193 cases in 1 year olds (28.5/100,000) and 111 cases in 2 year olds (16.4/100,000).'

Q6 - Setting and location
(show question description)

Question description: Is relevant contextual information (such as setting and location) provided that may influence the findings of the study?

Explanation: The manuscript provides contextual information about the setting and location that could influence study findings, specifically focusing on the UK, where the Bexsero vaccine has been introduced for infants and is being evaluated for broader use. This geographical focus is relevant as it informs the disease burden and policy context.

Quotes:

  • In September 2015 the UK became the first country in the world to routinely offer to infants a vaccine (Bexsero) against MenB disease at 2, 4 and 12 months of age.
  • The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.

Q7 - Comparators
(show question description)

Question description: Are the interventions or strategies being compared described, along with the rationale for their selection?

Explanation: The manuscript clearly describes the vaccination strategies compared, focusing on catch-up vaccination with Bexsero for various age groups, and provides the rationale for their selection based on economic attractiveness and declining disease burden as children age. The manuscript also details the criteria used to determine cost-effectiveness and the reasoning behind focusing on certain age groups.

Quotes:

  • The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.
  • Offering catch-up vaccination to increasingly older children is less economically attractive because of declining disease burden.
  • Providing catch-up vaccination to older birth cohorts is less economically attractive due to a decreasing disease burden, and even extending vaccine to 2 year olds could not be considered cost-effective.

Q8 - Perspective
(show question description)

Question description: What perspective(s) were adopted by the study, and why were they chosen?

Explanation: The study adopted the NHS and personal and social services perspective in its cost-effectiveness analysis. This perspective was chosen to align the analysis with typical health economic evaluations in the UK, considering the costs to the healthcare system.

Quotes:

  • "analyses were undertaken from the NHS and personal and social services perspective."

Q9 - Time horizon
(show question description)

Question description: What is the time horizon for the study, and why is it appropriate?

Explanation: The manuscript does not specify a fixed time horizon for the study. Instead, it uses a base case model that draws disease incidence and case fatality estimates from a seven-year period of past data (2005/06-2011/12) to consider future costs and benefits. This approach is explained in the 'Methods' section, but it does not establish a specific time horizon duration commonly seen in economic evaluations.

Quotes:

  • The base case model uses incidence over a seven year period to allow for such future changes in disease, but this is higher than the low burden currently experienced.
  • Disease incidence and case fatality estimates were drawn from Hospital Episode Statistics data over a seven year (2005/06-2011/12) period to allow for potential future increases in disease; in a historical context the UK is currently experiencing low disease incidence.

Q10 - Discount rate
(show question description)

Question description: What discount rate(s) were used, and what was the rationale for choosing them?

Explanation: The discount rates used in the study were 3.5% and 1.5% for future costs and benefits. The rationale for choosing these rates was to explore different scenarios of cost-effectiveness based on varying discounting rates, consistent with UK health economic evaluation guidelines, where 3.5% is a standard rate, and 1.5% represents a lower rate for sensitivity analyses.

Quotes:

  • Future costs and benefits were discounted at 3.5% in the base case and analyses were undertaken from the NHS and personal and social services perspective.
  • We also assessed the effect of using 1.5% discounting for cost and benefits, including family and network QALYs.
  • Conversely using highly favourable assumptions ... with 1.5% discounting.

Q11 - Selection of outcomes
(show question description)

Question description: What outcomes were used as measures of benefit and harm?

Explanation: The manuscript specifies the use of 'Quality Adjusted Life Years' (QALYs) as a measure of benefit, reflecting gains achieved through vaccination, such as reducing disease cases, sequelae, and death. It also considers costs due to adverse reactions as a measure of harm when evaluating the cost-effectiveness of the vaccination strategies.

Quotes:

  • The benefits of vaccination are captured through gains in Quality Adjusted Life Years (QALY) through reducing disease cases, sequelae and death.
  • Vaccine delivery costs are modelled separately from the cost of the vaccine and costs of adverse reactions are also considered.
  • Strategies were deemed cost-effective if the discounted cost per QALY gained was <$20,000 for the base case.

Q12 - Measurement of outcomes
(show question description)

Question description: How were the outcomes used to capture benefits and harms measured?

Explanation: The manuscript describes the outcomes used to measure both benefits and harms in terms of Quality Adjusted Life Years (QALYs). Benefits are captured through gains in QALYs by reducing disease cases, sequelae, and death, while the harms and costs are evaluated through the cost per QALY gained at various price points for the vaccine and the discounted cost-effectiveness ratios.

Quotes:

  • The benefits of vaccination are captured through gains in Quality Adjusted Life Years (QALY) through reducing disease cases, sequelae and death.
  • Strategies were deemed cost-effective if the discounted cost per QALY gained was <$20,000 for the base case.
  • In the base case model, catch-up vaccination of 1 year old children (83.8% uptake) with Bexsero could be considered cost-effective... with a threshold of $20,000 per QALY gained.

Q13 - Valuation of outcomes
(show question description)

Question description: What population and methods were used to measure and value the outcomes?

Explanation: The methods section of the manuscript describes the population targeted by the study and the model used to measure outcomes. It focuses on the cost-effectiveness of catch-up vaccination for children aged 1-4 years, using a transmission dynamic mathematical and economic model that incorporates various cost parameters, vaccine efficacy, and disease incidence data.

Quotes:

  • We adapted the transmission dynamic mathematical and economic model used to inform the infant vaccination JCVI decision to consider additional catch-up vaccination options.
  • The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.

Q14 - Measurement and valuation of resources and costs
(show question description)

Question description: How were the costs valued in the study?

Explanation: The manuscript specifies how costs were valued, including vaccine delivery, healthcare services, and consideration of QALY gains. It clearly outlines different factors such as the price per vaccine dose, healthcare costs, and long-term support costs in the economic model.

Quotes:

  • "The fee given to GPs for administering vaccines and immunisations increased from $7.50 to $9.80 per dose from 1 April 2016."
  • "The model includes costs of: acute hospital care and initial follow-up appointments, public health response, long term support for survivors with sequelae, and litigation claims against the NHS."
  • "Future costs and benefits were discounted at 3.5% in the base case and analyses were undertaken from the NHS and personal and social services perspective."

Q15 - Currency, price, date, and conversion
(show question description)

Question description: What are the dates of the estimated resource quantities and unit costs, and what currency and year were used for conversion?

Explanation: The manuscript specifies the dates for hospital episode statistics data used, which are 2005/06-2011/12. It indicates that costs and unit prices were considered for April 2016, and costs were in USD with comparisons to historical UK costs.

Quotes:

  • "Disease incidence and case fatality estimates were drawn from Hospital Episode Statistics data over a seven year (2005/06-2011/12) period..."
  • "The fee given to GPs for administering vaccines and immunisations increased from $7.50 to $9.80 per dose from 1 April 2016."

Q16 - Rationale and description of model
(show question description)

Question description: If a model was used, was it described in detail, including the rationale for its use? Is the model publicly available, and where can it be accessed?

Explanation: The manuscript provides a detailed description of the model used to evaluate the cost-effectiveness of catch-up MenB vaccination. It also discusses its adaptations and assumptions, such as transmission dynamics and vaccination schedules. The model is an adaptation of a previously used and reviewed model for infant vaccinations, implying public availability through prior publications.

Quotes:

  • "We adapted the transmission dynamic mathematical and economic model used to inform the infant vaccination JCVI decision to consider additional catch-up vaccination options."
  • "A strength of this work is that it builds upon a previously independently reviewed model which was used to inform the infant recommendation for the use of Bexsero in the UK."

Q17 - Analytics and assumptions
(show question description)

Question description: What methods were used for analyzing or statistically transforming data, extrapolation, and validating any models used?

Explanation: The manuscript outlines the statistical and modeling methods used for analysis, including the adaptation of previously used transmission dynamic mathematical and economic models, consideration of different vaccination strategies, and multiple sensitivity analyses. It also discusses parameter values, including discount rates for future costs and benefits, and evaluates cost-effectiveness using QALYs and incremental cost-effectiveness ratios (ICERs) at different vaccine price points.

Quotes:

  • We adapted the transmission dynamic mathematical and economic model used to inform the infant vaccination JCVI decision to consider additional catch-up vaccination options.
  • Future costs and benefits were discounted at 3.5% in the base case and analyses were undertaken from the NHS and personal and social services perspective.
  • Strategies were deemed cost-effective if the discounted cost per QALY gained was <$20,000 for the base case.

Q18 - Characterizing heterogeneity
(show question description)

Question description: What methods were used to estimate how the results vary for different sub-groups?

Explanation: The manuscript does not provide information on specific methods used to estimate how results vary for different sub-groups. It describes cost-effectiveness analyses with different assumptions related to vaccine efficacy, age cohorts, and discounting for costs and benefits, but it does not mention methods for estimating results for different sub-groups.

Quotes:

  • "The cost-effectiveness of catch-up vaccination was considered incrementally on the existing routine infant programme; we did not consider catch-up vaccination beyond 4 year olds because our previous work (data not shown) had indicated this would not be cost-effective."
  • "In the conservative scenario with 66% vaccine strain coverage and no herd effects (all other parameters at base case values), using a threshold of $30,000 per QALY gained, only catch-up vaccination in 1 year old children could be considered cost-effective and only with a very low vaccine price of <=$2 per dose (Table 2)."

Q19 - Characterizing distributional effects
(show question description)

Question description: How were the impacts distributed across different individuals, and were adjustments made to reflect priority populations?

Explanation: The manuscript does not provide a detailed description of how impacts were distributed across different individuals, nor does it mention adjustments made to reflect priority populations. It primarily focuses on the cost-effectiveness of vaccination strategies by age groups rather than addressing specific population adjustments.

Quotes:

  • "The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds."
  • "Based on the criteria currently used by JCVI our models suggest only catch-up vaccination in 1 year old children could be recommended on economic grounds, incremental to the existing infant programme, but only if the vaccine could be procured at a low cost."

Q20 - Characterizing uncertainty
(show question description)

Question description: What methods were used to characterize sources of uncertainty in the analysis?

Explanation: The manuscript does not provide specific details on methods used to characterize sources of uncertainty in the analysis. While it mentions input assumptions and scenario analyses, it does not detail specific approaches or statistical methods for uncertainty characterization like probabilistic sensitivity analysis or confidence intervals.

Quotes:

  • There remains, however a considerable degree of uncertainty around many of the model parameters and whilst surveillance data will help to reduce some of this uncertainty, it is still too early in the programme to be able to revise any of the assumptions, for example of vaccine effectiveness.
  • The base case model uses incidence over a seven year period to allow for such future changes in disease, but this is higher than the low burden currently experienced.

Q21 - Approach to engagement with patients and others affected by the study
(show question description)

Question description: Were patients, service recipients, the general public, communities, or stakeholders engaged in the design of the study? If so, how?

Explanation: The manuscript does not mention any engagement of patients, service recipients, the general public, communities, or stakeholders in the design of the study. The study design is primarily based on modelling of cost-effectiveness using existing data and assumptions, without reference to involvement from external parties.

Quotes:

  • In September 2015 the UK became the first country in the world to routinely offer to infants a vaccine (Bexsero) against MenB disease at 2, 4 and 12 months of age. The decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive disease and that immunising this group could be cost-effective if the vaccine was procured at a low price.
  • The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.

Q22 - Study parameters
(show question description)

Question description: Were all analytic inputs or study parameters (e.g., values, ranges, references) reported, including uncertainty or distributional assumptions?

Explanation: The manuscript includes details about various parameters used in the model, such as vaccine efficacy rates and costs, but it does not fully specify the uncertainty or distributional assumptions for these parameters, which is a key part of reporting all analytic inputs or study parameters comprehensively.

Quotes:

  • Our base case models assume the vaccine has some ability to disrupt transmission and carriage although the evidence for this is limited at present.
  • It is thought that carriage prevalence is low in young children, thus the potential herd effects generated through vaccinating this age group may not be large.
  • QALY gains in family and network members were considered in sensitivity analyses.

Q23 - Summary of main results
(show question description)

Question description: Were the mean values for the main categories of costs and outcomes reported, and were they summarized in the most appropriate overall measure?

Explanation: The manuscript reports mean values for costs and outcomes in terms of vaccine prices per dose required to achieve cost-effectiveness. These are summarized in terms of willingness-to-pay per QALY (Quality Adjusted Life Year), which is an appropriate overall measure for assessing cost-effectiveness.

Quotes:

  • In the base case model, catch-up vaccination of 1 year old children (83.8% uptake) with Bexsero could be considered cost-effective, incremental on the existent routine infant programme, if the vaccine could be procured at a low vaccine price, estimated at <=$8 per dose with a threshold of $20,000 per QALY gained (Table 1).
  • Reducing the discounting for costs and benefits from 3.5% to 1.5% improves the incremental cost-effectiveness ratio and increases the threshold vaccine price.
  • In this scenario catch-up vaccination in 1 year olds could be cost-effective at $20 per dose, incremental on the infant programme, extending this to 2 year olds and then to 3-4 year olds could be incrementally cost-effective at $12 and $6 per dose respectively.

Q24 - Effect of uncertainty
(show question description)

Question description: How did uncertainty about analytic judgments, inputs, or projections affect the findings? Was the effect of the choice of discount rate and time horizon reported, if applicable?

Explanation: The effect of the choice of discount rate and time horizon on the findings is reported in the manuscript. It discusses how altering the discount rate from 3.5% to 1.5% affects the cost-effectiveness outcomes, increasing the threshold vaccine price for catch-up vaccination.

Quotes:

  • Reducing the discounting for costs and benefits from 3.5% to 1.5% improves the incremental cost-effectiveness ratio and increases the threshold vaccine price.
  • In this scenario catch-up vaccination in 1 year olds could be cost-effective at $20 per dose, incremental on the infant programme, extending this to 2 year olds and then to 3-4 year olds could be incrementally cost-effective at $12 and $6 per dose respectively.

Q25 - Effect of engagement with patients and others affected by the study
(show question description)

Question description: Did patient, service recipient, general public, community, or stakeholder involvement make a difference to the approach or findings of the study?

Explanation: The manuscript does not mention any involvement of patients, service recipients, general public, community, or stakeholders influencing the study's approach or findings. The text focuses on modeling cost-effectiveness using mathematical and economic models, along with specific assumptions and economic analyses.

Quotes:

  • The decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive disease and that immunising this group could be cost-effective if the vaccine was procured at a low price.
  • The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.
  • Our model estimates that offering Bexsero to 1 year old children in a catch-up campaign could be cost-effective at $8 per dose with a willingness to pay of $20,000 per QALY.

Q26 - Study findings, limitations, generalizability, and current knowledge
(show questiondescription)

Question description: Were the key findings, limitations, ethical or equity considerations, and their potential impact on patients, policy, or practice reported?

Explanation: While the manuscript reports key findings on the cost-effectiveness of MenB vaccinations and acknowledges limitations and uncertainties, it does not explicitly cover ethical or equity considerations, nor does it discuss the potential impact on patients, policy, or practice beyond economic evaluations.

Quotes:

  • A strength of this work is that it builds upon a previously independently reviewed model which was used to inform the infant recommendation for the use of Bexsero in the UK. There remains, however a considerable degree of uncertainty around many of the model parameters.
  • The window of opportunity to vaccinate these individuals is also limited. Babies born after 1 May 2015 are already eligible for vaccination through the NHS infant schedule, thus the cohort of toddlers who are aged 1 and who have not been vaccinated is reducing.
  • Based on the criteria currently used by JCVI our models suggest only catch-up vaccination in 1 year old children could be recommended on economic grounds, incremental to the existing infant programme, but only if the vaccine could be procured at a low cost.

SECTION: TITLE
Modelling the cost-effectiveness of catch-up 'MenB' (Bexsero) vaccination in England

SECTION: ABSTRACT
Highlights

We modelled the cost-effectiveness catch-up 'MenB' vaccination in England.

Catch-up vaccination for 1 year old children could be cost-effective at =$8 per dose.

Additionally vaccinating 2 year olds was less cost-effective.

With conservative vaccine assumptions vaccinating 2 year olds was not-cost-effective.

It was not cost-effective to extend vaccination further to 3-4 year olds.

We assessed the cost-effectiveness of offering catch-up vaccination with Bexsero against meningococcal disease to children too old to receive the vaccine under the recently introduced infant programme. Offering catch-up vaccination to increasingly older children is less economically attractive because of declining disease burden. We estimate catch-up vaccination of 1 year old children could be cost-effective, incremental on the infant programme with a vaccine price of =$8 per dose. Extending vaccination to 2 year olds could only be cost-effective (incremental on infant and 1 year old catch-up) with a vaccine price of =$3 per dose and was not cost-effective in sensitivity analyses with more conservative vaccine assumptions. Extending catch-up further to 3-4 year olds was not cost-effective. Employing the current criteria for assessing vaccines, our models suggest that even with low vaccine prices only catch-up vaccination in 1 year old children could be cost-effective, when considered incrementally on the infant programme.

SECTION: INTRO
Introduction

In September 2015 the UK became the first country in the world to routinely offer to infants a vaccine (Bexsero) against MenB disease at 2, 4 and 12 months of age. The decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive disease and that immunising this group could be cost-effective if the vaccine was procured at a low price.In September 2015 the UK became the first country in the world to routinely offer to infants a vaccine (Bexsero) against MenB disease at 2, 4 and 12 months of age. The decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive diseaseThe decision to immunise infants was made by the Joint Committee on Vaccination and Immunisation (JCVI), based on evidence that this age group are the most at risk of invasive disease and that immunising this group could be cost-effective if the vaccine was procured at a low price.

In spring 2016 the largest health petition in UK history was received by parliament, calling for vaccination against meningococcal group B disease (MenB) for all children up the age of 11 years. As part of the original vaccine decision making process, there were several iterations of mathematical and economic models, which considered many different vaccination strategies, including catch-up strategies targeting pre-school (1-4 years) or school-aged (5-17 years) children. One of the key uncertainties is whether Bexsero can prevent transmission of the meningococcus and induce herd protection. Assuming that the vaccine provides direct protection only, our previous models have shown that catch-up vaccination was unlikely to be cost-effective, and therefore could not be recommended by JCVI.

The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.The aim of this modelling study was to further investigate the cost-effectiveness of different catch-up options, focusing not on children under 11 years, but on the birth cohorts after infancy who experience the greatest disease burden, i.e. 1, 2 and 3-4 year olds.

SECTION: METHODS
Methods

We adapted the transmission dynamic mathematical and economic model used to inform the infant vaccination JCVI decision to consider additional catch-up vaccination options.We adapted the transmission dynamic mathematical and economic model used to inform the infant vaccination JCVI decision to consider additional catch-up vaccination options. We modelled vaccination as a two dose schedule delivered 1 month apart in catch-up cohorts (vaccine uptake in catch-up cohorts was assumed to be the same as for the MenC vaccine campaign). In the base case we assumed the vaccine covered 88% of circulating meningococcal strains with a 30% vaccine efficacy against carriage acquisition and 95% vaccine efficacy against disease.

We used the same parameter values in the model as considered previously, except for the price for the vaccine delivery cost. The fee given to GPs for administering vaccines and immunisations increased from $7.50 to $9.80 per dose from 1 April 2016.. The model includes costs of: acute hospital care and initial follow-up appointments, public health response, long term support for survivors with sequelae, and litigation claims against the NHS. Disease incidence and case fatality estimates were drawn from Hospital Episode Statistics data over a seven year (2005/06-2011/12) periodDisease incidence and case fatality estimates were drawn from Hospital Episode Statistics data over a seven year (2005/06-2011/12) period to allow for potential future increases in disease; in a historical context the UK is currently experiencing low disease incidence. Vaccine delivery costs are modelled separately from the cost of the vaccine and costs of adverse reactions are also considered. The benefits of vaccination are captured through gains in Quality Adjusted Life Years (QALY) through reducing disease cases, sequelae and death; QALY gains in family and network members were considered in sensitivity analyses. Previously the JCVI specified a QALY adjustment factor of x3 should be applied in the model for long-term sequelae due to concerns surrounding the ability of current tools to adequately capture quality of life losses due to meningococcal disease. This adjustment factor was retained in this iteration of the model.

The cost-effectiveness of catch-up vaccination was considered incrementally on the existing routine infant programme; we did not consider catch-up vaccination beyond 4 year olds because our previous work (data not shown) had indicated this would not be cost-effective. Future costs and benefits were discounted at 3.5% in the base case and analyses were undertaken from the NHS and personal and social services perspective.Future costs and benefits were discounted at 3.5% in the base case and analyses were undertaken from the NHS and personal and social services perspective.d analyses were undertaken from the NHS and personal and social services perspective. Strategies were deemed cost-effective if the discounted cost per QALY gained was $20,000 for the base case. Strategies were deemed cost-effective if the discounted cost per QALY gained was $20,000 for the base case. For consistency with the routine infant immunisation decision by JCVI we considered a conservative scenario assuming 66% strain coverage for the vaccine and no herd effects (no vaccine efficacy against carriage acquisition); this was deemed cost-effective if the discounted cost per QALY gained was $30,000. We also assessed the effect of using 1.5% discounting for cost and benefits, including family and network QALYs, assuming a lower incidence and returning to the previous vaccine delivery cost of $7.50 per dose.

SECTION: RESULTS
Results

In the base case model, catch-up vaccination of 1 year old children (83.8% uptake) with Bexsero could be considered cost-effective, incremental on the existent routine infant programme, if the vaccine could be procured at a low vaccine price, estimated at =$8 per dose with a threshold of $20,000 per QALY gained (Table 1). Extending the catch-up to include 2 year olds (75.6% uptake) was less economically attractive, driven by the fact that disease incidence in 2 year olds is lower than in 1 year old children; in the absence of vaccination the model assumes 359 cases in infants (annual incidence 52.9/100,000 persons, 0 year olds), 193 cases in 1 year olds (28.5/100,000) and 111 cases in 2 year olds (16.4/100,000). We estimated that catch-up vaccination in 2 year olds, incremental on the routine infant programme and 1 year old catch-up, could only be cost-effective if the vaccine were priced $3 per dose or less. It was not possible to find a positive vaccine price when extending catch-up further to 3 and 4 year olds (annual incidence 11.2 and 8.4/100,000 respectively; 75.6% vaccine uptake in both year groups), and since disease incidence falls further after this age, the same applies up to age 11 years.

Reducing the discounting for costs and benefits from 3.5% to 1.5% improves the incremental cost-effectiveness ratio and increases the threshold vaccine price. In this scenario catch-up vaccination in 1 year olds could be cost-effective at $20 per dose, incremental on the infant programme, extending this to 2 year olds and then to 3-4 year olds could be incrementally cost-effective at $12 and $6 per dose respectively.

If the previous vaccine delivery cost of $7.50 per dose is used instead of the new $9.80 fee, the estimated 'cost-effective' vaccine prices for 1 and 2 year old catch-up are increased by $2.30 a dose and extending catch-up to 3-4 year olds could then be deemed cost-effective at a vaccine price =$2 per dose.

Disease incidence naturally varies over time even in the absence of intervention against MenB disease. The base case model uses incidence over a seven year period to allow for such future changes in disease, but this is higher than the low burden currently experienced. Reducing the modelled number of annual cases of disease by a third, to more closely resemble the incidence experienced currently, rules out many catch-up strategies from an economic perspective. Only 1 year old catch-up could be cost-effective in this scenario and only with a vaccine price of =$1 per dose.

In the conservative scenario with 66% vaccine strain coverage and no herd effects (all other parameters at base case values), using a threshold of $30,000 per QALY gained, only catch-up vaccination in 1 year old children could be considered cost-effective and only with a very low vaccine price of =$2 per dose (Table 2). However, if family and network QALYs were included this could be increased to $6 per dose. Under extremely conservative assumptions (lower disease incidence, 66% strain coverage, no herd effects or litigation costs) none of the catch-up policies could be considered cost-effective with 3.5% discounting. Conversely using highly favourable assumptions (91% strain coverage, 60% vaccine efficacy against carriage, including litigation costs and quality of life losses in family and network members, with 1.5% discounting) catch-up vaccination of 1, 2 and 3-4 year olds could be incrementally cost-effective at $35, $24 and $18 per dose respectively with a willingness to pay of $20,000 per QALY gained.

SECTION: DISCUSS
Discussion

Our model estimates that offering Bexsero to 1 year old children in a catch-up campaign could be cost-effective at $8 per dose with a willingness to pay of $20,000 per QALY. Providing catch-up vaccination to older birth cohorts is less economically attractive due to a decreasing disease burden, and even extending vaccine to 2 year olds could not be considered cost-effective.

A strength of this work is that it builds upon a previously independently reviewed model which was used to inform the infant recommendation for the use of Bexsero in the UK. There remains, however a considerable degree of uncertainty around many of the model parametersA strength of this work is that it builds upon a previously independently reviewed model which was used to inform the infant recommendation for the use of Bexsero in the UK. There remains, however a considerable degree of uncertainty around many of the model parameters and whilst surveillance data will help to reduce some of this uncertainty, it is still too early in the programme to be able to revise any of the assumptions, for example of vaccine effectiveness. To our knowledge this is the first study to consider the cost-effectiveness of extending catch-up vaccination in individual birth cohorts after infancy. Previous models have either considered routine programmes only or catch-up in a number of birth cohorts combined.

Although our models suggest catch-up vaccination in 1 year olds could be cost-effective, it may be challenging to achieve this in practice. Our base case models assume the vaccine has some ability to disrupt transmission and carriage although the evidence for this is limited at present. It is thought that carriage prevalence is low in young children, thus the potential herd effects generated through vaccinating this age group may not be large. However, excluding any effect on carriage reduces the already low cost-effective vaccine price. The price procured per dose of vaccine for the infant programme is confidential. Thus while the price per dose for 1 year old catch-up appears similar to that estimated for the infant programme, if additional factors had to be included for a vaccine price to be agreed for the infant programme (such as the removal of an infant meningococcal C conjugate vaccine dose from the current schedule), procurement of doses for catch-up may not be possible at the prices we have suggested. Changes have also recently been made to the vaccine delivery cost and while there is currently no scope for reverting to the previous cost, our findings highlight that vaccine recommendation decisions can be affected by the administration fee. The window of opportunity to vaccinate these individuals is also limited. Babies born after 1 May 2015 are already eligible for vaccination through the NHS infant schedule, thus the cohort of toddlers who are aged 1 and who have not been vaccinated is reducing. Given the seasonal nature of meningococcal disease any catch-up vaccination would need to be done sufficiently early to afford protection to these children before the winter peak in disease to maximise the benefit from immunisation.

SECTION: CONCL
Conclusions

Based on the criteria currently used by JCVI our models suggest only catch-up vaccination in 1 year old children could be recommended on economic grounds, incremental to the existing infant programme, but only if the vaccine could be procured at a low cost.

SECTION: TABLE
Cost-effectiveness of catch-up vaccination assuming 88% vaccine strain coverage, 30% vaccine efficacy against carriage acquisition and 95% vaccine efficacy against disease.

Vaccine strategy Vaccination strategy compared with no vaccination Catch-up vaccination incremental on previous (row above) strategya ICER at $75/vaccine doseb ICER at $75/vaccine doseb Vaccine price at $30 k/QALYc Vaccine price at $20 k/QALYc 3.5% discounting for costs and benefits 2,4 + 12 months $168,000 - - - 2,4 + 12 months + CU in 1 y $167,400 $143,200 $13 $8 2,4 + 12 months + CU in 1-2 y $167,900 $199,800 $7 $3 2,4 + 12 months + CU in 1-4 y $170,100 $264,800 $2 NP : : 1.5% discounting for costs and benefits 2,4 + 12 months $114,200 - - - 2,4 + 12 months + CU in 1 y $113,600 $79,000 $29 $20 2,4 + 12 months + CU in 1-2 y $113,600 $111,000 $19 $12 2,4 + 12 months + CU in 1-4 y $114,100 $147,800 $12 $6

ICER, incremental cost-effectiveness ratio; QALY, Quality Adjusted Life Year; CU, catch-up vaccination; NP, positive vaccine price not possible.

2,4 + 12 months + CU in 1 y incremental on 2,4 + 12 months; 2,4 + 12 months + CU in 1-2 y incremental on 2,4 + 12 months + CU in 1 y etc.

Figures rounded to nearest 100.

Figures rounded to nearest $1.

Cost-effectiveness of catch-up vaccination assuming 66% vaccine strain coverage, 0% vaccine efficacy against carriage acquisition and 95% vaccine efficacy against disease.

Vaccine strategy Vaccination strategy compared with no vaccination Catch-up vaccination incremental on previous (row above) strategya ICER at $75/vaccine doseb ICER at $75/vaccine doseb Vaccine price at $30 k/QALYc Vaccine price at $20 k/QALYc 3.5% discounting for costs and benefits 2,4 + 12 months $273,400 - - - 2,4 + 12 months + CU in 1 y $273,100 $262,700 $2 NP 2,4 + 12 months + CU in 1-2 y $274,800 $401,800 NP NP 2,4 + 12 months + CU in 1-4 y $280,300 $613,700 NP NP : : 1.5% discounting for costs and benefits 2,4 + 12 months $188,600 - - - 2,4 + 12 months + CU in 1 y $188,100 $151,100 $11 $6 2,4 + 12 months + CU in 1-2 y $188,500 $233,000 $4 NP 2,4 + 12 months + CU in 1-4 y $190,300 $358,300 NP NP

ICER, incremental cost-effectiveness ratio; QALY, Quality Adjusted Life Year; CU, catch-up vaccination; NP, positive vaccine price not possible.

2,4 + 12 months + CU in 1 y incremental on 2,4 + 12 months; 2,4 + 12 months + CU in 1-2 y incremental on 2,4 + 12 months + CU in 1 y etc.

Figures rounded to nearest 100.

Figures rounded to nearest $1.