Public health impact and return on investment of the pediatric immunization program in Poland

ABSTRACT

Background

This study aims to evaluate the epidemiological impact and return on investment of the pediatric immunization program (PIP) in Poland from the healthcare-sector and societal perspectives.

Research design and methods

A health-economic model was developed focusing on the nine vaccines, targeting 11 pathogens, recommended by the public health authorities for children aged 0–6 years in Poland. The 2019 birth cohort (388,178) was followed over their lifetime, with the model estimating discounted health outcomes, life-years gained, quality-adjusted life-years, and direct and indirect costs with and without the PIP based on current and pre-vaccine – era disease incidence estimates, respectively.

Results

Across 11 targeted pathogens, the Polish PIP prevented more than 452,300 cases of disease, 1,600 deaths, 37,900 life-years lost, and 38,800 quality-adjusted life-years lost. The PIP was associated with vaccination costs of €54 million. Pediatric immunization averted €65 million from a healthcare-sector perspective (benefit–cost ratio [BCR], 2.2) and averted €358 million from a societal perspective (BCR, 7.6). The BCRs from both perspectives remained >1.0 in scenario analyses.

Conclusions

The Polish PIP, which has not previously been systematically assessed, brings large-scale prevention of disease-related morbidity, premature mortality, and associated costs. This analysis highlights the value of continued investment in pediatric immunization in Poland.

KEYWORDS:

• Vaccination
• benefit–cost ratio
• model
• pediatric
• return on investment

1. Introduction

Immunization programs are among the most successful and cost-effective public health strategies, preventing estimated 4 to 5 million premature deaths worldwide each year [1]. Vaccines dramatically reduce the burden of infectious disease and associated morbidity, mortality, and disability [2]. However, published evidence may not fully reflect the benefits of routine immunization programs (e.g. reductions in long-term complications due to infectious disease, productivity losses for caregivers, and quality-of-life impairments associated with disease cases) [3]. Because disease cases prevented by routine immunization programs may be unrecognized and can be challenging to quantify, the public health impact of these programs is often undervalued. To demonstrate the value of and ensure continued investment in routine immunization programs, it is important to evaluate specific countries’ return on investment (ROI) of these programs.

Nine vaccines recommended by the National Institute of Public Health (Narodowy Instytut Zdrowia Publicznego Państwowy Zakład Higieny – Państwowy Instytut Badawczy; NIZP-PZH) for children aged 0–6 years are included in Poland’s pediatric immunization schedule: diphtheria, tetanus, and whole-cell pertussis (DTwP); Haemophilus influenzae type b (Hib); hepatitis B; inactivated poliovirus vaccine (IPV); measles, mumps, rubella (MMR); pneumococcal conjugate vaccine (PCV); rotavirus; diphtheria, tetanus, and acellular pertussis (DTaP); and Bacille Calmette-Guérin (BCG). Rotavirus immunization, introduced in 2021, is the most recent addition to Poland’s immunization schedule. The routine vaccination program in Poland is regulated by law and includes mandatory immunizations.

Prior analyses have estimated the costs and economic impact of routine immunization in children in various geographies. Immunization funding represents a small proportion of total healthcare spending, with a median of 0.3% of the spending dedicated to immunization across 28 European countries, and 0.2% in Poland [4]. The costs of immunization are heavily outweighed by cost savings from averted disease morbidity and mortality. In the United States (US), for a single birth cohort, the routine childhood immunization program yielded savings of $13.5–$13.7 billion in direct health-care costs and of $55.1-$68.8 billion to society, with benefit-cost ratios of 2.8–3.0 and 7.5–10.1, respectively [5,6]. In a similar study conducted in Belgium, for the 2020 birth cohort, the program yielded savings of €35 million and €268 million in direct health-care costs and societal costs, respectively, and benefit-cost ratios of 1.4 and 3.2, respectively [7]. As of April 2023, the broad economic value of the pediatric immunization program (PIP) in Poland has not previously been evaluated in the scientific literature. The objective of this analysis was to evaluate the epidemiological impact and ROI of the PIP in Poland from both the healthcare-sector and societal perspectives.

2. Patients and methods

2.1. Model description

We developed a decision tree model in Microsoft Excel (Microsoft, Redmond, WA) to analyze the impact of Poland’s PIP. The model structure, which was used to analyze the US PIP, has previously been described in Carrico et al. [5]. The model focused on eight vaccines targeting 11 pathogens in the PIP for children aged ≤6 years. Due to its recent adoption in 2021, there are limited data on the impact of rotavirus vaccine in Poland. Therefore, rotavirus was included only in the scenario analysis. Vaccination of the 2019 Polish birth cohort was modeled using Poland’s pediatric immunization schedule recommended by NIZP-PZH and vaccination coverage estimates (Figure 1; Table 1). Outcomes were calculated over the birth cohort’s lifetime.

Figure 1. Poland’s pediatric immunization program, 2021 schedule.

BCG = Bacille Calmette-Guérin; DTaP = diphtheria, tetanus, and acellular pertussis; DTwP = diphtheria, tetanus, and whole-cell pertussis; HepB = hepatitis B; Hib = Haemophilus influenzae type b; IPV = inactivated poliovirus; MMR = measles, mumps, rubella; PCV = pneumococcal conjugate 10-valent vaccine.

Note: This reflects Poland’s pediatric immunization program schedule as of January 2021 [8].

Table 1. Childhood vaccine coverage estimates and vaccine public price.

Vaccine	Vaccination coverage (% fully vaccinated)^a	Public price per dose^b
DTwP	85.3%	€6.97
HepB	90.9%	€2.41
Hib	85.2%	€5.99
MMR	86.1%	€5.94
PCV-10	84.3%	€17.89
IPV	77.4%	€6.62
Rotavirus^c	95.2%	€8.20
BCG	98.0%	€2.91
DTaP^d	77.6%	€12.30

BCG = Bacille Calmette-Guérin; DTaP = diphtheria, tetanus, and acellular pertussis; DTwP = diphtheria, tetanus, and whole-cell pertussis; HepB = hepatitis B; Hib = Haemophilus influenzae type b; IPV = inactivated poliovirus; MMR = measles, mumps, rubella; PCV-10 = pneumococcal conjugate 10-valent vaccine.

^aAll vaccination coverage estimates were obtained from NIZP-PZH [9]. DTwP and Hib vaccine coverage for the first three doses was based on data from the 2017 birth cohort and dose 4 coverage was based on 2016 birth cohort. HepB vaccine coverage was based on data from the 2017 birth cohort. MMR vaccine coverage for the first dose was based on data from the 2016 birth cohort and dose 2 coverage was calculated using data from the 2008 birth cohort. PCV-10 vaccine coverage for dose 1 and 2 was based on data from the 2017 birth cohort and dose 3 coverage was calculated using data from the dose 3 and 4 DTwP vaccine coverage. IPV vaccine coverage for dose 1 and 2 was based on data from the 2017 birth cohort, dose 3 coverage was calculated using data from the 2016 birth cohort, and dose 4 was calculated using data from the 2012 birth cohort. BCG vaccine coverage was based on 2018 birth cohort. DTaP vaccine coverage was based on data from the 2012 birth cohort.

^bAll vaccine acquisition costs were obtained from MZ [10] as of October 2020.

^cRotavirus was included only in scenario analysis. Because the rotavirus vaccine has been recommended only since 2021, Poland coverage data were not available. Therefore, the model assumed that rotavirus vaccine coverage for all three doses was the same as DTwP coverage for the first three doses.

^dDTaP at 6 years is a booster recommended for the whole population.

Two analytical scenarios were compared: one in which routine pediatric immunization occurred according to Poland’s pediatric immunization schedule and the other in which no immunization occurred. In the scenario with immunization, incidence rates of the modeled diseases reflected current incidence rates, whereas incidence rates in the scenario without immunization reflected pre-vaccine rates. This analytic approach captures both the direct and indirect effects of vaccination, as pre-vaccine and vaccine-era incidence rates are at the population level (i.e. vaccine-era incidence was among both vaccinated and unvaccinated individuals). The model then calculated the difference in lifetime health outcomes and costs between the PIP scenario and a scenario without any immunization. Separate decision trees were used to estimate disease cases and costs associated with each pathogen targeted in Poland’s PIP. Specifically, for each disease, outcomes were calculated for the scenarios without and with vaccination for each month from birth until 12 months and then annually thereafter. For the birth cohort’s lifetime, we subtracted the number of individuals who died (due either to all-cause or to disease-specific mortality) from the cohort for that period. The remaining individuals progressed to the next period. Age-specific all-cause mortality, disease incidence, treatment costs per case by disease severity, and probability of different outcomes by disease severity were used to parameterize the decision tree for each modeled disease.

Analyses were conducted from both a healthcare-sector perspective and a societal perspective. Costs from the healthcare-sector perspective included medical costs incurred by the Polish health system, except for costs of capitated care (i.e. care that is reimbursed at a fixed amount per patient over a period of time rather than based on healthcare resource utilization), which are not typically included in Polish health technology assessments. The societal perspective additionally included productivity losses. Health outcomes and costs were discounted at an annual rate of 3.5% and 5.0%, respectively, according to the Polish recommendations for economic evaluations [11]. Cost data were converted from 2020 local currency (Polish złoty, PLN) to 2020 Euros using an exchange rate of 0.225 [12].

2.2. Vaccination program costs

We calculated the costs of the vaccination program based on the 2020 immunization schedule and used the most recent vaccine coverage data available (Table 1), which were for the 2014–2018 period. Vaccination coverage for each dose was multiplied by the number of children in the birth cohort alive at each age when vaccination was recommended to calculate the number of vaccine doses administered. Children were assumed to receive each vaccine at the earliest recommended age for each dose.

Vaccine acquisition costs were obtained from public prices from the Department of Public Procurement at the Minister of Health [10]. Costs associated with a wastage rate of 0.2% of vaccine doses were applied based on responses from Poland’s Chief Sanitary Inspectorate. Vaccine administration costs were excluded from the analysis because costs of outpatient care in Poland are incurred from capitated payments, which were excluded from the analysis. Vaccine-related adverse events have incurred costs associated with specialist outpatient visits and hospitalizations [13–15].

2.3. Disease incidence

The analysis used incidence data before and after each vaccine was routinely recommended in Poland for each vaccine-preventable disease (Table 2). Pre-vaccine incidence was obtained from published incidence estimates for Poland, and analyses conducted by the Polish Agency for Health Technology Assessment and Tariffication or were calculated using annual published case estimates and Poland population data for the same period [16–43]. Pre-vaccine incidence rates from other European countries were used for Poland when local data were unavailable, which included Belgium for pneumonia outpatient visits due to Streptococcus pneumoniae (S. pneumoniae) and Central Europe for acute otitis media due to S. pneumoniae [37,38,44].

Table 2. Pre-vaccine and vaccine-era disease incidence estimates.

Pathogen	Disease incidence per 100,000 by age group^a						Sources
	<1 y	1–4 y	5–9 y	10–19 y	20–64 y	≥65 y
Diphtheria							[9,16–18,35,36,45,46]
Without PIP (pre-vaccine)	120	120	120	120	120	120
With PIP (vaccine era)	0	0	0	0	0	0
Hepatitis B							[9,21,23,34–36,45]
Without PIP (pre-vaccine)	26	19	2	2	10–25	25
With PIP (vaccine era)	0.055	0	0	0–0.040	0.076–0.187	0.187
H. influenzae type b							[9,20–22,34–36,45]
Without PIP (pre-vaccine)^b	4	1–3	0.263	0.057–0.058	0.017–0.057	0.043
With PIP (vaccine era)	0.152	0–0.143	0	0	0–0.002	0.001
Measles							[9,24,25,34–36,45]
Without PIP (pre-vaccine)	859	3,313	1,847	24–253	0.948–8	0.948
With PIP (vaccine era)	2	2–3	0.850	0.369–0.564	0.150–0.494	0
Mumps							[9,34–36,45,47–51]
Without PIP (pre-vaccine)	40	102–392	961	58–386	2–10	0.582
With PIP (vaccine era)	3	10–33	34	8–18	0.879–3	0.337
Pertussis							[52,53]
Without PIP (pre-vaccine)	664	739–1,041	408	10–89	10	10
With PIP (vaccine era)	44	29–49	24	6–44	6	6
Polio							[9,35,36,39,45,46]
Without PIP (pre-vaccine)	12	12	12	12	12	12
With PIP (vaccine era)	0	0	0	0	0	0
Rotavirus^b							[54–60]
Hospitalizations^c
Without PIP (pre-vaccine)	370	370	NA	NA	NA	NA
With PIP (vaccine era)	70	122	NA	NA	NA	NA
Outpatient visits^c
Without PIP (pre-vaccine)	2,960	2,960	NA	NA	NA	NA
With PIP (vaccine era)	813	1,184	NA	NA	NA	NA
Non – medically attended cases^c
Without PIP (pre-vaccine)	11,840	11,840	NA	NA	NA	NA
With PIP (vaccine era)	2,384	2,384	NA	NA	NA	NA
Rubella							[9,34–36,45,47–51]
Without PIP (pre-vaccine)	200	110–329	897	62–382	1–11	0
With PIP (vaccine era)	45	29–44	29	8–11	0.380–5	0.034
S pneumoniae							[31–38,42,43,61–64]
Invasive pneumococcal disease
Without PIP (pre-vaccine)	5	3–5	0.873	0.083–0.277	0.177–2	4
With PIP (vaccine era)	1	1	0.655	0.063–0.208	0.143–2	4
Pneumonia hospitalizations
Without PIP (pre-vaccine)	6,709	1,688	381	84–381	31–179	690
With PIP (vaccine era)	5,863	1,475	306	61–306	22–128	644
Acute otitis media
Without PIP (pre-vaccine)^d	25,453	30,427	11,433	0–11,433	NA	NA
With PIP (vaccine era)^d	11,376	13,598–17,132	6,756	0–6,756	NA	NA
Tetanus							[16,65–68]
Without PIP (pre-vaccine)	0	0.680	2	0.620–2	0.360–1	1
With PIP (vaccine era)	0	0	0	0	0.010–0.024	0.098
Tuberculosis							[41]
Without PIP (pre-vaccine)	174	174	174	174–256	368–371	238
With PIP (vaccine era)	1	1	1	1–3	10–24	20

NA = not applicable; PIP = pediatric immunization program.

^aRange indicates that incidence varied by age group within the presented range. Incidence values that were less than 1 per 100,000 people were reported to three decimals. Values shown as 0 are true zeros (i.e. no observed cases in the age group).

^bRotavirus was included only in scenario analysis because rotavirus immunization was introduced only in 2021.

^cIncidence was not modeled for ages 5+ years.

^dIncidence was not modeled for ages 18+ years.

Our analysis assumed that any observed reduction in disease incidence after vaccine introduction was considered fully attributable to vaccination within the population. This approach of using pre-vaccine and vaccine-era incidence captured the impact that vaccine efficacy, coverage, and waning have on disease incidence at the population level.

Vaccine-era incidence for each disease was obtained from recent surveillance data (2013–2019) from the NIZP-PZH – Państwowy Instytut Badawczy or from published incidence estimates for Poland, or were calculated by applying a reduction in disease incidence between the post- and pre-vaccine eras [9,35,36,41,45,53,55–61,65–71]. Vaccine-era incidence rates for Poland were calculated using data from other European countries when local data were unavailable. Specifically, data from Finland were used for S. pneumoniae, as Finland and Poland use the same type of PCV vaccine [69].

For both pre-vaccine and vaccine-era incidence, age-specific incidence was used when available. Invasive pneumococcal disease (IPD) pre-vaccine incidence was adjusted by an underreporting factor of 3.7 [37]. All other disease underreporting was assumed to be captured in the incidence estimates if the underlying data source is adjusted for underreporting. Additional details regarding the date of vaccination program initiation for each disease, as well as sources for pre-vaccine and vaccine-era incidence estimates, are available in Supplementary Material Table S1.

2.4. Healthcare-Sector disease Costs

To capture the disease-related costs, the model considered disease-specific case severity distributions, disease case-fatality rates, direct medical costs per case (by severity), and either one-time costs or annual costs for management of long-term sequelae, if applicable (Table 3). Costs of hospitalized cases were estimated based on Polish diagnosis-related group system (JGP system). If a given disease entity is treated under several codes, the average cost is estimated. Hospitalization costs were calculated based on unit costs provided by Zarządzenie Prezesa NFZ or obtained from published literature [13,15]. Costs of outpatient cases were estimated assuming that they will be treated in the primary care by a general practitioner. The cost of visits to a general practitioner or emergency department were not included in the analysis because costs of outpatient care in Poland are incurred from capitated payments. The cost of reimbursed drugs for outpatient cases were included [10,37,72]. Costs from the previous years were inflated to 2020 PLN using the Polish consumer product index [12,18].

Table 3. Severity of disease outcomes and medical costs per outcome.

Pathogen	Percent of cases	Cost per outcome (€)
Diphtheria
Hospitalized	100.0%	411
Outpatient case	0.0%	0
Hepatitis B
Fulminant case	0.1–0.6%	769
Hospitalized case	95.0–95.5%	886
Outpatient case (jaundice)	4.4%	0
H. influenzae type b
Meningitis case	93.1% −17.2%	1,128
Epiglottitis case	0.0%	338
Bacteremia case	5.8% − 51.0%	1,207
Pneumonia case	0.6% − 22.0%	955
Cellulitis case	0.0%	475
Arthritis case	0.0%	880
Other invasive case	0.5% − 9.9%	530
Measles
Encephalitis case	0.1% − 0.2%	1,021–1,128
Pneumonia case	1.6% − 7.2%	955
Otitis media case	1.4% − 13.5%	292–419
Uncomplicated or diarrhea case	79.2% − 96.1%	201–263
Mumps
Complicated case	1.5%	679–785
Uncomplicated case	98.5%	0
Pertussis
Hospitalized case	26.2%	706
Outpatient case	73.8%	2^a
Non – medically attended case	0.0%	0
Polio
Paralytic case	90.5%	1,021
Nonparalytic case	9.5%	0
Rotavirusb
Hospitalized case	─	624
Outpatient case	─	0
Non – medically attended case	─	0
Rubella
CRS	0.2%^c	2,207
Complicated case	0.6%	679–865
Uncomplicated	99.2% − 99.4%	0
S. pneumoniae
Invasive pneumococcal disease^d
Meningitis	─	2,390–2,554
Bacteremia	─	1,868–1,870
S. pneumoniae pneumoniad
Hospitalized case	─	690–797
Outpatient case	─	0
S. pneumoniae acute otitis media
Complex case	41.0%	21–24
Simple case	59.0%	1–5
Tetanus
Hospitalized case	100.0%	530
Outpatient case	0.0%	0
Tuberculosis
Hospitalized case	93.3%	4,117
Outpatient case	6.7%	637e

CRS = congenital rubella syndrome; NA = not applicable.

Note: A range indicates that a parameter varied by age group and/or pre or vaccine-era period within the presented range. Values and sources for all parameters presented in this table are presented in additional detail in the Supplementary Material.

^aIncludes the cost of reimbursed antibiotics (azithromycin or clarithromycin) [10,72].

^bRotavirus was included only in scenario analysis. Incidence rates for each specific rotavirus health outcome (including hospitalizations, outpatient visits, and non – medically attended cases) are presented in Table 2.

^cAll CRS cases are among those aged less than 1 year.

^dIncidence rates for invasive pneumococcal disease and S. pneumoniae pneumonia are presented in Table 2.

^eIncludes the cost of a 1-day hospitalization for diagnostic purposes, followed by ambulatory treatment comprising three specialist visits and antibiotic therapy [13,14,73–75].

The costs for the management of long-term sequelae for hepatitis B, IPD, measles, rubella, polio, and tuberculosis were obtained from published literature from Poland [13,76–79]. The annual cost of permanent paralysis was taken from a Belgium study and was converted to Polish current currency and then converted to 2020 Euros, due to a lack of local data [18,80,81].

2.5. Productivity losses

The human capital approach was applied to calculate the value of time loss due to acute disease, long-term complications, and disease-related mortality [6,82,83]. To calculate productivity losses due to acute disease, mean workdays lost per disease case (Supplementary Material Table S17) were multiplied by daily productivity, calculated from gross domestic product using methods described in Golicki et al. [84] and using data from the Central Statistical Office of Poland [18]. To calculate productivity losses associated with long-term complications and disease-related deaths, age-specific annual productivity was derived from daily productivity estimates [18] and Polish life expectancy estimates [85] discounted over time (Supplementary Material Table S20). The percentage reduction in annual productivity for long-term complications was assumed to be equal to the percentage reduction in health-related utility weights associated with the complication (Supplementary Material Table S18). The cost of caregiver time and travel for vaccination were not included in the analysis, consistent with local health technology assessment guidelines [11], and productivity losses due to vaccine-related adverse events were also not included as a simplifying assumption.

2.6. Quality-of-life impacts

Individuals who experienced vaccine-preventable disease or vaccine-related adverse events had quality-of-life impacts. To calculate quality-adjusted life-years (QALYs) lost from disease cases and adverse events, the number of disease cases and adverse events experienced by the birth cohort is multiplied by case- and event-specific disutilities and their associated durations (Supplementary Material Tables S3 and S16). For long-term complications, quality-of-life decrements were applied for the duration of the complication (i.e. number of years).

2.7. Model outcomes

Vaccine acquisition costs associated with the Polish PIP were calculated by multiplying the number of vaccine doses administered by the public price per dose. Vaccine-related adverse-event costs were calculated by multiplying adverse-event incidence per dose, number of vaccine doses administered, and cost per adverse event.

The number of cases for the 11 modeled pathogens for the 2019 birth cohort was calculated using incidence rates from the pre-vaccine and vaccine eras. Health outcomes by severity were then calculated, including the number of cases with long-term sequelae and disease-related deaths. The total number of cases of the disease is calculated each year; the values were then discounted annually and aggregated over the lifetime of the 2019 birth cohort.

Costs of disease for the 11 modeled pathogens were calculated for the scenario with the current PIP and for the scenario without the PIP where incidence rates returned to pre-vaccine levels. Costs of disease were calculated by multiplying the severity-specific cost per disease outcome by the number of clinical outcomes in the pre-vaccine and vaccine eras. Costs of long-term sequelae were calculated using annual costs, discounted over the duration of the sequela or over the cohort’s remaining lifetime. For the societal perspective only, productivity losses for acute cases, long-term sequelae, and disease-related deaths were included.

QALYs were calculated by multiplying the disutility and duration value per disease or adverse-event outcome and the number of clinical outcomes. The monetary value of total QALYs gained was calculated by multiplying the total QALYs gained by a willingness-to-pay threshold of €34,913 per QALY gained, which reflected the official willingness-to-pay threshold for health technology assessments in Poland in 2020 [86].

2.8. Analyses

2.8.1. Base-case analysis

The financial benefit–cost ratio (BCR) of the Polish PIP was calculated for each perspective by dividing the discounted lifetime costs of disease cases averted by the net vaccination costs, as done in similar previous analyses [5,6]. Health outcomes averted, specifically cases and deaths, due to the PIP were also calculated. QALYs gained were calculated as the difference in QALYs lost between the scenarios with and without the Poland PIP. The monetary value of QALYs gained was not included in the financial BCR for the base-case analysis.

2.8.2. Scenario analyses

In addition to the base-case analysis, we conducted a scenario analysis (1) in which routine rotavirus vaccination was included in Poland PIP. Incidence of rotavirus was limited to ages 0–4 years, given lack of data in older age groups in the pre-vaccine period and as the majority of the clinical burden of rotavirus is experienced by young children [87]. Annual rotavirus vaccination coverage in Poland was assumed to be the same as the DTwP vaccine coverage for the first three doses [9]. Vaccine acquisition costs for the rotavirus vaccine were obtained from public prices from the Department of Public Procurement at the Minister of Health [10]. Pre-vaccine incidence was obtained from estimates of the rotavirus burden of disease in European Union countries, and reductions in medically attended and non – medically attended rotavirus cases due to routine rotavirus immunization were estimated from data from Austria and Belgium, respectively [54–60] (Table 2). Details for disease-related inputs are presented in the Supplementary Material.

We conducted additional scenario analyses to assess the robustness of model results in changes in key assumptions. Scenarios considered the following relative variations from base-case input values: (2–3) 10% and 20% reduction in pre-vaccine disease incidence; (4–5) 10% and 20% increase in vaccine-era disease incidence; (6–7) 10% increase and decrease in vaccine acquisition costs; (8–9) 20% increase and decrease in healthcare-sector disease-related costs; (10) 20% reduction in case-fatality rates for all diseases; (11) discount rates set to 0%; (12) inclusion of the economic value of QALYs gained in the benefits variable of the financial BCR calculation; and (13) no underreporting adjustment to IPD incidence (3.7 underreporting factor in base-case analysis) [37].

3. Results

3.1. Discounted health outcomes without and with vaccination

For the 2019 Polish birth cohort of 388,178 individuals, without the PIP, there were estimated 549,600 preventable disease cases, resulting in 1,800 disease-related deaths 39,300 life-years (LYs) lost, and 40,500 QALYs lost because of disease-related morbidity and mortality over the cohort’s lifetime. When modeled with the Polish PIP, there were estimated 97,300 vaccine-preventable disease cases and 104 disease-related deaths, with 1,400 LYs and 1,700 QALYs lost. Therefore, the Polish PIP was associated with approximately 452,300 disease cases averted, 1,600 disease-related deaths averted 37,900 LYs gained, and 38,800 QALYs gained.

Figure 2 presents incremental health outcomes associated with the Polish PIP. The PIP’s impact on morbidity and mortality varied by disease, with the most discounted cases averted for S. pneumoniae (420 cases of IPD 11,600 cases of pneumococcal pneumonia, and 256,100 cases of acute otitis media averted), measles (79,100 cases averted), and tuberculosis (26,400 cases averted). Disease-related deaths averted, LYs gained, and QALYs gained were highest for diphtheria (830 deaths averted 19,600 LYs gained, and 18,000 QALYs gained), pertussis (230 deaths averted, 6,200 LYs gained, and 6,100 QALYs gained), and tuberculosis (210 deaths averted, 4,800 LYs gained, and 4,800 QALYs gained) (Figure 2). The PIP resulted in a > 90% reduction in disease cases for 10 of the targeted pathogens, with a ≥ 99% reduction in cases of measles, hepatitis B, tetanus, diphtheria, and polio.

Figure 2. Impact of the Poland PIP on health outcomes: base-case results by disease.

PIP = pediatric immunization program; QALY = quality-adjusted life-year.

Note: Figure a is the number of cases averted by the PIP. Figure b is the number of premature deaths averted by the PIP. Figure c is the life-years gained by the PIP. Figure d is QALYs gained by disease. All health outcomes were discounted at an annual rate of 3.5% [11].

3.2. Discounted economic outcomes without and with vaccination

Without the Polish PIP, lifetime societal disease-related costs for the birth cohort were €425 million, with 45% (€191 million) owing to productivity losses from disease-related mortality, 29% (€124 million) owing to healthcare-sector costs to treat cases of disease, and the remainder owing to productivity losses from cases of disease and long-term sequelae (€110 million). With the Polish PIP, lifetime societal disease-related costs were reduced to €13 million, in which €5.0 million correspond to healthcare-sector costs to treat cases of disease, €4.4 million are attributed to productivity losses due to cases of disease and long-term sequelae, and €3.4 million are due to productivity losses due to disease-related mortality. Without the Polish PIP, tuberculosis contributed the majority (60%) of healthcare-sector disease-related costs (€75 million); whereas with the Polish PIP, S. pneumoniae contributed the majority (58%) of healthcare-sector costs for managing disease cases (€2.9 million). Societal disease-related cost savings due to Polish PIP totaled €412 million, with the highest savings coming from averted cases of diphtheria, tuberculosis, and measles (Table 4). For disease-related costs averted, healthcare-sector costs to treat individuals with acute cases of disease and long-term sequelae accounted for 29% of the savings, while the remaining costs averted were from productivity losses due to cases of disease and long-term sequelae (26%) and productivity loss due to mortality (45%) (Table 4).

Table 4. Return on investment for the Poland PIP compared with no PIP.

Incremental results	Healthcare-sector perspective (€ millions)	Societal perspective (€ millions)
Incremental vaccination costs	54	54
Acquisition	54	54
Adverse events	0.2	0.2
Disease-related costs averted	119	412
Disease treatment	119	119
Productivity loss due to disease	^─	106
Productivity loss due to disease-related mortality	^─	187
Total costs averted	65	358
Financial benefit-cost ratio	2.2	7.6
Value of QALYs gained^a	^─	1,356

PIP = pediatric immunization program; QALY = quality-adjusted life-year.

Note: Costs are presented in 2020 Euros discounted at an annual rate of 5%.

^aThe value of QALYs gained is calculated by multiplying the total QALYs gained (38,830) with the PIP by a willingness-to-pay threshold of €34,913 per QALY gained [86].

Immunization of the 2019 Polish birth cohort resulted in €54 million in vaccination costs, with nearly all costs associated with vaccine acquisition. Adverse events constitute a small proportion of costs (<1%). Vaccination costs were substantially outweighed by disease-related costs averted (€412 million), resulting in net savings of €358 million and a financial BCR of 7.6 from the societal perspective. This BCR indicates that every €1 invested in the Polish PIP is expected to result in over €7 in savings to society. When productivity losses were excluded, net savings for the healthcare-sector perspective amounted to €65 million, and a financial BCR of 2.2 was observed. Although the economic value of QALYs gained because of the Polish PIP was not included in the financial BCR calculation in the base-case analysis, the economic value of total QALYs gained was approximately €1.4 billion when a willingness-to-pay threshold of €34,913 per QALY gained was considered.

3.3. Scenario analyses

When the impact of routine rotavirus vaccination was included in the scenario analysis (Scenario 1), 36900 rotavirus health-care visits (4,500 hospitalizations 32,400 outpatient visits) and 165,100 non – medically attended cases were averted, resulting in an additional 590 QALYs gained, and €17 million societal disease-related costs averted (societal BCR for Polish PIP = 6.8).

Scenarios related to the input data demonstrated that BCRs were most impacted when a 0% discount rate was applied (BCR = 36.5; Scenario 11) and when pre-vaccine incidence for all diseases was reduced by 20% (BCR = 6.0; Scenario 3) (Figure 3). When the economic value of QALYs gained was included in the BCR calculation, the societal BCR increased substantially compared with the base-case analysis (BCR = 32.6; Scenario 12). BCRs from the healthcare-sector and societal perspectives remained above 1 in all scenarios.

Figure 3. Results for scenarios considering variations in key input values.

BCR = benefit–cost ratio; PIP = pediatric immunization program.

4. Discussion

We estimated that 452,300 disease cases and 1,600 premature deaths were averted for the 2019 Polish birth cohort over its lifetime due to pediatric immunization, resulting in net savings both from a societal perspective (BCR = 7.6) and from a healthcare-sector perspective (BCR = 2.2). As a result of the PIP, the incidences of diphtheria, hepatitis B, H influenzae type b, polio, and tetanus were reduced to <1 case per 100,000 population annually and a reduction of >90% in incidence for 10 of the 11 pathogens targeted by the PIP. When key input values were varied in scenario analyses, financial BCRs from the healthcare-sector and societal perspectives remained above 1, highlighting the robustness of the ROI of the Polish PIP.

While routine rotavirus immunization has only been in place in Poland since 2021, scenario analysis was conducted assuming similar reductions in disease activity following routine immunization introduction as observed in Austria and Belgium. Scenario results estimated significant reductions in rotavirus-related health-care visits (36,900) and non-medically attended cases (165,100). The BCR of the Polish PIP was slightly lower in the scenario analysis (6.8) relative to the base-case analysis (7.6). Real-world reductions in rotavirus disease from routine immunization in Poland will likely be dependent on vaccination coverage rates achieved and local epidemiological factors, although significant reductions in rotavirus-related healthcare use have been observed across European countries with routine rotavirus immunization despite variations in vaccination coverage rates [87].

To our knowledge, this is the first analysis to systematically assess the public health and economic effects of Poland’s PIP. Previous ROI analyses that evaluated the US PIP reported BCRs ranging from 2.8 to 3.0 and 7.5 to 10.1 from healthcare-sector and societal perspectives, respectively [5,6]. Estimated BCRs for this Polish analysis were similar to those estimated in the US. BCR differences across countries could be explained by differences in the timing of immunization program implementation for each vaccine, vaccines and dosing regimens used in immunization programs, magnitude and consistency of vaccination uptake, disease epidemiology, and health-care costs, among other factors. Nevertheless, results from this study and previous research demonstrate the value and importance of continued investment in national immunization programs and maintenance of high vaccination coverage rates. This research focuses on pediatric immunization programs, and as the European population ages, life-course immunization strategies will become even more important to ensure that the pediatric, adolescent, and adult populations are sufficiently protected against infectious disease.

A key strength of this analysis was that our analytic framework captured population-level reductions in incidence from the pre-vaccine period to today due to both direct and indirect effects of vaccination without having to model direct protection (using estimates of vaccine efficacy, waning, and vaccine coverage) and indirect protection of vaccination such as herd immunity, serotype replacement, and shifts in age of infection. However, the benefits of herd immunity are likely underestimated because avoided cases in older ages were heavily discounted [5].

This analysis had limitations. First, other public health improvements beyond pediatric immunization that have occurred in parallel with vaccine introductions over time were not directly accounted for in the analysis. It is widely accepted that lower incidences of pediatric infectious diseases are largely attributable to vaccine introduction within the population [88–92], although other improvements in Poland’s public health system and hygiene may have also contributed to reduced disease activity and burden. To demonstrate the influence of this assumption, we conducted a multi-way scenario analysis in which the base-case incidence values for each of the 11 vaccine-preventable pathogens were decreased simultaneously by 10%, as a proxy for testing a smaller change in incidence specifically attributable to vaccination.

Further, the analysis did not consider the costs of vaccines administered to adolescent and adult populations, specifically adult PCV and Td/Tdap boosters, and the proportions of disease reduction attributable to these vaccines were not considered. As a result, there is the potential for overestimation of the proportion of the impacts on morbidity, mortality, and economic burden associated with vaccine-preventable disease that is attributable to the pediatric immunization program. However, the extended benefits of vaccination, including improved health equity, increased resiliency of health systems, and reduced use of antimicrobials for vaccine-preventable diseases were not considered [93]. The impact and value of pediatric immunization may have been underestimated as a result.

Vaccination costs in our analysis did not include the costs of vaccine delivery (e.g. transportation and cold-chain storage), although costs associated with vaccine wastage were incorporated. While limited data were available to estimate vaccine delivery costs in high-income countries, a recent systematic literature review estimated a median delivery cost (including supply chain costs, startup costs, and recurrent costs) of $2.64 per dose (2016 USD) for delivering a schedule of vaccines in low- and middle-income countries [94]. The value of the PIP was robust to assumed increases in vaccine acquisition costs in scenario analyses, suggesting that inclusion of vaccine delivery costs would not change the findings of the analysis. Additionally, the analysis also considered public vaccine prices, which are higher than tendered prices. Scenario analyses that considered hypothetical reductions in vaccine prices improved the PIP’s ROI, indicating that the use of public prices underestimates the PIP’s value for money.

Consistent with previous economic evaluations of immunization programs, we used the human capital approach to estimate productivity losses due to mortality [5,6,95,96]. While European economic evaluation guidelines discuss the societal perspective, they do not specify what approach should be used for valuing productivity losses and gains [97]. A recent systematic literature review of cost-effectiveness analyses for vaccines found that 16 of 24 studies reporting their methods for productivity estimation used the human capital approach, with the remaining studies applying the friction cost approach [98]. Use of the friction cost approach for productivity loss estimation would result in reduced productivity losses due to mortality. However, BCRs greater than 1.0 were observed in analyses from the healthcare-sector perspective, where productivity losses were excluded, demonstrating that the ROI of the Polish PIP was robust to assumptions regarding productivity losses.

Data limitations must also be noted. Specifically, for diseases that have been mostly eradicated (e.g. diphtheria and polio), current estimates of health outcomes (e.g. case-fatality ratios), rates of long-term complications, and disease treatment costs were limited. Disease outcome and cost data were obtained from Polish data sources and published economic analyses as available. For example, acute costs of severe diphtheria and paralytic polio cases were estimated from Polish hospitalization costs for diagnostic-related group codes associated with similar syndromes. When Polish data are unavailable, data from other countries and/or assumptions guided by subject matter experts are applied. Scenario analyses to explore variations in assumptions for disease incidence, case-fatality ratios, and healthcare-sector disease treatment costs revealed that the PIP’s ROI remained robust to this uncertainty. As a conservative approach, disease-underreporting factors were not applied except for IPD; thus, the disease impact and value of pediatric immunization may have been underestimated. Finally, while limitations in the analysis varied in the impact they were expected to have on the estimated value of pediatric immunization, returns on investment estimates from both economic perspectives were robust to broad variations in key model input values.

5. Conclusions

This analysis estimated that Poland's PIP averts substantial morbidity and mortality, with estimated 452,300 cases of the disease and 1,600 premature deaths prevented for the 2019 Polish birth cohort. Additionally, the Polish PIP contributes cost savings from the healthcare-sector and societal perspectives due to reductions in cases of disease and disease-related costs. Every €1 invested in the PIP returned more than €2 to health-care payer, and more than €7 in savings to society (i.e. from a health-care and societal perspective, respectively). Continued investment in the PIP is needed to sustain the public health and economic benefits of pediatric immunization.

Declaration of interest

C Mellott, J Carrico, S Talbird, and M Clinkscales are employed (or were employed at the time of the analysis) by RTI Health Solutions, which received funding from Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., and Rahway, NJ, U.S.A., for the conduct of this study. R Jaworski is an employee of MSD Polska Sp and owns stock in Merck & Co., Inc., Rahway, NJ, U.S.A. G Bencina is an employee of MSD Spain and owns stock in Merck & Co., Inc., Rahway, NJ, U.S.A. E Karamousouli is an employee of MSD Greece, and owns stock in Merck & Co., Inc., Rahway, NJ, U.S.A. A Eiden is an employee of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, and owns stock in Merck & Co., Inc., Rahway, NJ, U.S.A. U Sabale is an employee of MSD Lithuania, and owns stock in Merck & Co., Inc., Rahway, NJ, U.S.A. I Dobrowolska was employed by HealthQuest Sp. z o.o., Warsaw, Poland, at the time of study. D Golicki is employed by HealthQuest Sp. z o.o., Warsaw, Poland, and is a shareholder in HealthQuest Sp. z o.o., Warsaw, Poland, which received funding from MSD Polska Sp. z o.o. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

CEM participated in the conception, design, and planning of the study; in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. RJ participated in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. JC participated in the conception, design, and planning of the study; in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. SET participated in the conception, design, and planning of the study; in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. ID participated in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. DG participated in the acquisition and analysis of the data; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. GB participated in the conception, design, and planning of the study; in the interpretation of the results; and in the drafting of the manuscript and critically reviewing and revising it for important intellectual content. MC participated in the conception, design, and planning of the study; in the acquisition and analysis of the data; in the interpretation of the results; and in the critically reviewing and revising of the manuscript for important intellectual content. EK participated in the interpretation of the results and in critically reviewing and revising the manuscript for important intellectual content. AE participated in the interpretation of the results and in critically reviewing and revising the manuscript for important intellectual content. The US participated in the interpretation of the results and in critically reviewing and revising the manuscript for important intellectual content. All authors gave final approval for the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for the content of the work.

Previous presentations

This work was presented in preliminary form as: Mellott C, Jaworski R, Carrico J, Clinkscales M, Talbird S, Dobrowolska I, Golicki D, Tsoumani E, Sabale U. Public health impact and return on investment of the pediatric immunization program in Poland. Poster presented at the ISPOR Europe 2022; 6 November 2022. Vienna, Austria

Ethical approval and patient consent

This study did not directly involve any human participants; therefore, ethical review and approval in accordance with the local legislation and institutional requirements was not required nor was informed consent required.

Acknowledgments

Kate Lothman of RTI Health Solutions provided medical writing support for the development of this manuscript. These services were funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14760584.2023.2275712

Additional information

Funding

The manuscript was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.