Standardized country-level delivery unit cost estimates for routine childhood, routine adolescent, and campaign vaccination: an updated modeling analysis

This study utilizes Bayesian meta-regression models on updated empirical data to generate standardized, country-level estimates of immunization delivery unit costs for routine childhood, routine adolescent, and campaign vaccination strategies across low- and middle-income countries, providing essential evidence for global vaccination planning and budgeting.

The Gist

Imagine the world's health system as a massive, global delivery service. Its job isn't to drop off pizza, but to deliver something even more important: vaccines.

This paper is essentially a price tag calculator for that delivery service.

Here is the simple breakdown of what the researchers did, why it matters, and what they found, using some everyday analogies.

1. The Problem: "We Need to Know the Price of the Truck, Not Just the Pizza"

For years, we've known how much the vaccines (the "pizza") cost to buy. But health officials often forget to calculate the cost of the delivery (the "truck," the "driver," the "gas," and the "warehouse").

• The Reality: In many poorer countries, the cost to deliver a vaccine is actually higher than the cost of the vaccine itself.
• The Gap: Many countries don't have up-to-date receipts. They are trying to budget for a trip without knowing the price of gas or the driver's salary. Some countries are trying to deliver vaccines to teenagers (a new route) or during emergency "campaigns" (a special express delivery), but they have no data on how much those specific trips cost.

2. The Solution: The "Global Costing GPS"

The researchers decided to build a GPS for costs. Instead of asking every single country to drive the route and measure the gas themselves (which takes years and costs a fortune), they looked at all the existing "trip reports" (studies) from the last 20 years.

• The Data: They gathered 318 "trip reports" from 20 countries.
• The Math: They used a fancy computer model (a "Bayesian meta-regression," which sounds scary but is just a very smart way of averaging things) to predict the cost for every country in the world, even those that never sent in a report.
• The Analogy: Think of it like a weather app. You don't need a thermometer in your exact backyard to know if it's raining; the app looks at data from nearby cities and your location's geography to give you a highly accurate forecast. This paper did the same thing for vaccine delivery costs.

3. The Three Routes They Mapped

The researchers calculated the "delivery fee" for three different types of trips:

1. The Daily Commute (Routine Childhood Vaccines): This is the standard delivery to babies and toddlers.

   • The Cost: On average, it costs about $5.86 to deliver one dose to a child in a developing country.
   • The Twist: It's cheaper in big, busy cities (economies of scale) and more expensive in remote villages where the driver has to drive over bumpy roads to reach just a few families.

2. The School Bus Run (Routine Adolescent Vaccines): This is a new route, targeting teenagers (specifically for HPV vaccines).

   • The Cost: This is the most expensive trip, averaging $17.65 per dose.
   • Why? It's harder to find teenagers. You can't just drop a vaccine at a baby's house; you have to go to schools or organize special clinics. It's like trying to deliver a package to a teenager who is constantly moving around, rather than a baby who stays home.

3. The Emergency Express (Campaigns): These are "sprint" deliveries, like a measles outbreak response where you vaccinate everyone in a week.

   • The Cost: Surprisingly, this is the cheapest per dose, averaging $3.13.
   • Why? Because you are moving fast and hitting a huge number of people at once. It's like a flash sale where the shipping cost per item drops because you are shipping 10,000 items in one truck.

4. The Big Surprise: Rich vs. Poor isn't Always What You Think

You might think it costs less to deliver vaccines in the poorest countries because labor is cheaper. The study found the opposite.

• The "Small Town" Tax: In very poor, remote, or small countries, the cost per dose is actually higher. Why? Because the health system is fragile. The roads are bad, the trucks break down, and the staff have to walk miles to reach a single village.
• The "Big City" Discount: In middle-income countries with bigger populations and better roads, the cost per dose is lower. They have "economies of scale." It's like a pizza delivery: if you live in a dense city, the driver drops off 50 pizzas in one hour. If you live in a remote farm, the driver drives 2 hours to drop off one pizza. The cost per pizza is much higher for the farm.

5. Why This Matters (The "So What?")

This paper is like a menu for the world's health budget.

• For Governments: It helps them plan their budgets. If they want to start a new vaccine program for teenagers, they can look at this paper and say, "Okay, we need to set aside about $18 per person for delivery."
• For Donors: It helps groups like Gavi (the Vaccine Alliance) decide where to send money. If a country is trying to do a campaign but doesn't have the cash for the "truck and driver," this data proves they need extra help.
• For the Future: As the world tries to reach more people (including older kids and adults), we need to know the price of the "delivery" to make sure the program doesn't run out of money halfway through.

The Bottom Line

This study didn't just count the price of the vaccine; it counted the price of the effort to get it into someone's arm. It tells us that while delivering vaccines to babies is getting cheaper in some places, reaching teenagers and remote villages remains a costly challenge. By having these numbers, the world can finally stop guessing and start planning with a clear map in hand.

Technical Summary

1. Problem Statement

Reliable estimates of immunization delivery costs are critical for health system planning, budgeting, and economic evaluation. However, significant evidence gaps persist, particularly for:

• Low- and Middle-Income Countries (LMICs): Many lack up-to-date, accurate delivery unit cost data.
• Specific Modalities: While routine childhood vaccination costs are somewhat documented, data for routine adolescent vaccination and mass campaign vaccination are scarce.
• Currency and Time: Existing data often relies on outdated studies or different currency years, hindering current policy decisions in a constrained fiscal environment.

The study aims to fill these gaps by producing updated, standardized country-level estimates for all LMICs across three delivery modalities: routine childhood, routine adolescent, and campaign vaccination.

2. Methodology

The authors employed a Bayesian meta-regression framework to synthesize empirical data and generate predictive models.

Data Sources:

• Primary Data: The 2024 update of the Immunization Delivery Cost Catalogue (IDCC), maintained by the Immunization Costing Action Network (ICAN).
• Scope: Included 32 studies for routine childhood (141 observations), 13 studies for routine adolescent (62 observations, using HPV vaccination as a proxy), and 34 studies for campaign vaccination (104 observations).
• Cost Definition: Focused on delivery costs (labor, supply chain, capital, and other service delivery), explicitly excluding vaccine product costs, injection supplies, and wastage.
• Perspectives: Distinguished between Financial costs (actual monetary outlays) and Economic costs (financial costs + opportunity costs, e.g., volunteer labor).

Model Specification:

• Statistical Approach: A Bayesian meta-regression model using a Generalized Linear Model (GLM) with a Gamma distribution and a log link function.
• Covariates: The model incorporated country-level covariates (log GDP per capita, population size, DTP3 coverage, under-five mortality, population density, urbanization) and study-level characteristics (cost category, financial vs. economic perspective, full vs. incremental costing).
• Model Selection: Best-fit models for each modality were selected using the minimized Watanabe-Akaike Information Criterion (WAIC).
• Implementation: Estimated using R (v4.4.1) and Stan (v2.32.6) via adaptive Hamiltonian Monte Carlo (4 chains, 5,000 iterations each).
• Output: Generated population-weighted average economic and financial cost per dose estimates for 2024 (in 2024 USD) for 136 LMICs, stratified by World Bank income levels.

3. Key Contributions

• Expansion of Scope: This is the first study to provide standardized, model-based estimates specifically for routine adolescent and campaign vaccination delivery across all LMICs, moving beyond the previous focus on routine childhood immunization.
• Updated Evidence Base: Leveraged the 2024 IDCC update, incorporating significantly more recent data than previous iterations.
• Methodological Rigor: Adopted a robust Bayesian framework with WAIC-based model selection to handle heterogeneity in study designs and cost categories.
• Granular Estimates: Provided country-specific estimates for 136 LMICs, allowing for tailored budgeting and strategic planning.

4. Key Results

Global Averages (2024 USD, Population-Weighted):

• Routine Childhood Vaccination:
  • Economic Cost: $5.86 per dose (95% UI: $2.74–$13.43).
  • Financial Cost: $3.02 per dose (95% UI: $1.52–$6.29).
• Routine Adolescent Vaccination (HPV Proxy):
  • Economic Cost: $17.65 per dose (95% UI: $7.76–$44.30).
  • Financial Cost: $10.08 per dose (95% UI: $4.15–$25.01).
  • Note: Adolescent costs are significantly higher, likely due to school-based delivery complexities and lower economies of scale compared to infant programs.
• Campaign Vaccination:
  • Economic Cost: $3.13 per dose (95% UI: $2.03–$4.78).
  • Financial Cost: $1.79 per dose (95% UI: $1.11–$2.91).

Drivers of Cost Heterogeneity:

• Routine Childhood: Costs decrease with larger population size, higher DTP3 coverage, and lower under-five mortality (indicating economies of scale and mature health systems).
• Adolescent: Costs are negatively associated with population density (higher density = lower costs), suggesting school-based delivery is more efficient in concentrated areas.
• Campaigns: Costs are significantly higher in smaller, lower-density, lower-income, and higher-mortality settings, reflecting the logistical challenges of reaching dispersed populations.

Income Level Disparities:

• Counter-intuitively, Low-Income Countries (LICs) often showed higher per-dose costs for routine childhood and adolescent vaccination compared to Lower-Middle-Income Countries (LMICs). This is attributed to LMICs achieving higher service volumes and better health system integration, allowing fixed costs to be spread over more doses.

5. Significance and Limitations

Significance:

• Policy Support: Provides a practical, evidence-based input for cost-effectiveness analyses and strategic planning where primary data is missing.
• Resource Allocation: Helps donors and governments (e.g., Gavi) understand the financial requirements for scaling up adolescent and campaign immunization under the Immunization Agenda 2030 (IA2030).
• Standardization: Offers a consistent methodology for comparing costs across different countries and modalities.

Limitations:

• Correlational Nature: Associations between predictors and costs are correlational, not causal, due to potential unobserved study characteristics.
• Data Sparsity: The model relies on a limited number of primary studies (especially for adolescent and campaign data), which may not fully capture within-country heterogeneity.
• Proxy Usage: Adolescent costs are proxied by HPV vaccination, which may differ in delivery strategy (e.g., school-based vs. facility-based) from other adolescent vaccines.
• Assumption of Uniformity: The model assumes delivery costs do not vary systematically by specific vaccine product (e.g., oral vs. injectable), which may not hold true in practice.

In conclusion, this study provides a critical, updated resource for global health stakeholders, highlighting that while campaign delivery is generally cheaper per dose than routine delivery, the structural and logistical context of a country significantly dictates the actual cost, with low-resource settings often facing higher per-unit delivery costs due to inefficiencies and lack of scale.