Estimating the money flow in the economy attributed to rotavirus disease and vaccination in the Netherlands using a Social Accounting Matrix (SAM) framework

University of Groningen

Estimating the money flow in the economy attributed to rotavirus disease and vaccination in

the Netherlands using a Social Accounting Matrix (SAM) framework

Kotsopoulos, Nikolaos; Haitsma, Gertruud; Connolly, Mark P.; Standaert, Baudouin

Expert review of pharmacoeconomics & outcomes research DOI:

10.1080/14737167.2020.1693269

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Kotsopoulos, N., Haitsma, G., Connolly, M. P., & Standaert, B. (2020). Estimating the money flow in the economy attributed to rotavirus disease and vaccination in the Netherlands using a Social Accounting Matrix (SAM) framework. Expert review of pharmacoeconomics & outcomes research, 20(6), 603-612. https://doi.org/10.1080/14737167.2020.1693269

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ierp20

Expert Review of Pharmacoeconomics & Outcomes

Research

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ierp20

Estimating the money flow in the economy

attributed to rotavirus disease and vaccination in

the Netherlands using a Social Accounting Matrix

(SAM) framework

Nikolaos Kotsopoulos , Gertruud Haitsma , Mark P. Connolly & Baudouin

Standaert

To cite this article: Nikolaos Kotsopoulos , Gertruud Haitsma , Mark P. Connolly & Baudouin Standaert (2020) Estimating the money flow in the economy attributed to rotavirus disease and vaccination in the Netherlands using a Social Accounting Matrix (SAM) framework, Expert Review of Pharmacoeconomics & Outcomes Research, 20:6, 603-612, DOI: 10.1080/14737167.2020.1693269

To link to this article: https://doi.org/10.1080/14737167.2020.1693269

© 2019 GlaxoSmithKline Biologicals S.A. Published by Informa UK Limited, trading as Taylor & Francis Group.

View supplementary material

Published online: 22 Nov 2019. _{Submit your article to this journal}

Article views: 1276 _{View related articles}

ORIGINAL RESEARCH

Estimating the money flow in the economy attributed to rotavirus disease and

vaccination in the Netherlands using a Social Accounting Matrix (SAM) framework

a_{Health Economics, Global Market Access Solutions Sarl, St-Prex, Switzerland;}b_{Department of Economics, University of Athens, Athens, Greece;} c_{Unit of Pharmacoepidemiology & Pharmacoeconomics, Department of Pharmacy, University of Groningen, Groningen, The Netherlands;}d_Research

& Development, GSK, Wavre, Belgium

ABSTRACT

Background: The economics of rotavirus gastroenteritis in infants <5 years old is well-known within healthcare. The financial consequences for families, employers and authorities are not so well explored. The present study evaluates how vaccine prevention changes money flows among those involved in the management of disease, and its consequences.

Methods: A Social Accounting Matrix (SAM) framework has been developed reflecting the distribution of income and spending at equilibrium affected by rotavirus disease among all those concerned for 1 year. The data came from official sources and published literature. A comparison of the financial equilibrium between with and without a national rotavirus immunization program has been conducted, along with sensitivity analysis for the results.

Results: The total financial cost difference at equilibrium between presence and absence of rotavirus vaccination was +€26.758 million over one year as a net economic surplus. The payment of vaccination (€19.194 million) by the government was offset by the increase in tax revenue (€14.561 million) and by the lower spending in treatment care (€7.998 million).

Conclusion: Studying the financial flows between different transacting agents can demonstrate the financial burden of a disease and the benefits of its prevention on agents’ income and spending.

ARTICLE HISTORY

Received 4 April 2019 Accepted 12 November 2019

KEYWORDS

Social accounting matrix; broader economics; public economics; rotavirus; vaccination

1. Introduction

Rotavirus gastroenteritis is a highly prevalent and contagious disease associated with substantial morbidity, mainly occur-ring in children under the age of 5 years old in those countries that do not include rotavirus vaccination in their national immunization program [1,2]. The Netherlands is still such a country and the number of rotavirus cases per year averages around 73,000 cases, including several fatalities [3,4]. Common symptoms are diarrhea, vomiting and fever, which urge par-ents to take time off work or to organize external caregiving for their sick child [5].

The human rotavirus vaccine (HRV,Rotarix, GSK) has been launched in 2006, and the vaccine effect has demonstrated to be consistent with the efficacy data obtained through clinical trials [6]. Several health-economic analyses confirmed that rotavirus vaccination is cost-effective in high income countries at an acceptable price for the vaccine [7]. But this health economic analysis often limits the evaluation from the per-spective of the healthcare budget holder being the Ministry of Health (MoH). This restricted view ignores possible conse-quences that health and disease can have on influencing different agents of the overall economy who interact with

each other [8]. Maybe surprisingly the most impactful element of the cost burden of rotavirus disease isn’t within the health-care delivery but is outside the healthhealth-care system caused by work absenteeism of one of the parents taking care for their sick child. Only 2 to 3% of the birth cohort will ever be exposed to hospital care for this disease up to the age of 5 years and maximum 10 to 15% will go for a 1stline medical visit. However, 40% of the children will need some care from their parents who may have to take days off work [9]. We are interested in understanding what those 40% of non-medical care mean regarding cost and effect they cause on the overall economy. In addition, we want to assess the cost changes that may happen when we inject extra money to avoid the disease through vaccine prevention [10].

To evaluate the overall influence of technology on the econ-omy, broadening the perspective of evaluations than only look at healthcare-related impacts is needed [11–13]. Secondly, investi-gation is needed for economic consequences outside healthcare, caused either by the disease or by health improvements. Such consequences include work absenteeism, reduced work presen-teeism, sick leave payment, productivity loss, less payment of wages, less income, less spending because of less income, and therefore less tax revenue [14–16]. These consequences imply

CONTACTNikolaos Kotsopoulos nikos@gmasoln.com Global Market Access Solutions Sarl, St-Prex, Switzerland

Prior congress activities: preliminary results were presented at EEROVAC (2019) European Expert Meeting on Rotavirus Vaccination - 6th Biennial (Apr 23-25, 2019, Riga, Latvia).

This article has been republished with minor changes. These changes do not impact the academic content of the article. Supplementary data for this article can be accessedhere.

EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 2020, VOL. 20, NO. 6, 603–612

https://doi.org/10.1080/14737167.2020.1693269

© 2019 GlaxoSmithKline Biologicals S.A. Published by Informa UK Limited, trading as Taylor & Francis Group.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

that the monetary condition of both individuals and public institutions are impacted by disease and disease prevention and treatment. Interestingly, all these effects can be expressed into monetary units which facilitates their evaluation potential. Thirdly, integrating the interconnected nature of health and the economy using a single analytic framework is needed.

One approach often used by economists to assess the effect of specific policies or interventions on the overall economy is the Social Accounting Matrix (SAM). SAM makes links at macro- and micro-economic levels of accounts of labor markets and house-holds to evaluate financial, economic, and social policies [17]. SAMs record money flows taking place within the overall econ-omy during limited timeframes, often one year. They were first developed in the 1960s and have been used ever since to assess economy-wide multipliers addressing poverty and issues of income distribution in developing countries. They are often developed as a basis for computable general equilibrium models [8,18]. Such a framework can simulate and adapt resource alloca-tion decisions for optimizing economic outcomes.

In our analysis we assess whether a SAM-model can demon-strate that seasonal rotavirus disease cases, influence the money flows in the overall economy of the Netherlands and to what extent extra-investment in prevention of the disease generates benefits to the economy that justify the investment. The pro-posed SAM framework encompasses the distribution of income and spending of those likely to be impacted by the disease [19]. It will explore how spending on rotavirus vaccination of birth cohorts in the Netherlands, will influence money flows among different economic agents. The key influential parameter is the impact of the disease on the productivity of the affected popula-tion which in turn, influences money flows to and from other agents. The underlying assumption is that healthier children should keep their parents at work resulting in higher productiv-ity, leading to better tax revenues for the authorities and less spending in healthcare treatment which should pay off the pay-ment of the vaccine by the authorities.

2. Methods

2.1. SAM framework

To demonstrate money flows relevant to microeconomic agents like households with children under the age of 5 years in the Netherlands, a SAM secondary distribution of income account (sub-model) was developed. The constructed SAM captures the impact of rotavirus on household labor income in contrast with a full SAM approach which includes a variety of unrelated economic transactions. This partial SAM focused on a limited number of agents and on money transactions directly relevant to the research question. Agents are domains over which para-meters, variables and equations are defined. An agent can also have sub-categories as needed depending on the question to be answered. The model starts by illustrating the financial con-sequences of an annual rotavirus epidemic among different economic agents considered as the endowment or ‘starting money condition’. The model then compares the ‘starting money condition’ for one year -named as condition A-, with an alternative condition in which the MoH has introduced universal mass vaccination (UMV) against rotavirus for a long time, there-fore yielding a new monetary equilibrium. The latter condition is also assessed for one year and is defined as condition B [17]. Consistent with the SAM approach, the model does not estimate how the money flows over time, rather it compares two money equilibrium conditions changed by an intervention being the UMV against rotavirus.

The structure of a SAM model is a square matrix as presented in

Figure 1. The matrix shows the list of agents selected in the model with their financial transactions or money flows between them. Each agent has both a column as a row account. The column account, by convention, demonstrates the spending or expendi-ture while the row account records the income or revenue. The agents selected in the SAM conditions A and B are firms, house-holds, government, health insurance companies (HICs), and vac-cine manufacturers. Households and government include subcategories defined by the type of working contract for the household, and by institution i.e. MoH and Ministry of Finance (MoF) for the government. Each cell in the SAM-matrix can simul-taneously disclose an expenditure by an agent and a source of income to another transacting agent. The attractiveness of work-ing with SAM models is that they are closed monetary systems meaning that total spending must equal total income. This also means that the system cannot lose or generate money when being in condition A or B. The total money flow circulating in condition A and B (cell G7 in the matrix) can only be different, when comparing the absence (condition A) to the presence of vaccination (condition B) and the consequential different eco-nomic dynamics generated by vaccine-induced prevention.

2.2. Model structure

The type and direction of the modeled money flows have been kept simple and follow a simple narrative that is described below. Firms pay wages to households and corporate taxes to the MoF (cells A2; A3 in Figure 1). Households consume parts of their income by purchasing from firms (B1), pay value-added tax (VAT) on goods and services consumed, income taxes (B3), and pay

Article Highlights

● Rotavirus disease causes outbreaks that seriously affect children below 5 years of age, in countries that do not implement vaccination programs. The Netherlands is one of these countries.

● Much of the disease burden goes beyond the health care system attributed to indirect costs (i.e. parents’ absenteeism) which are not captured in conventional economic evaluations

● We evaluate the extended impact of rotavirus outbreaks across different sectors of the economy applying a Social Accounting Matrix (SAM) which measures transactions and money flows across different economic sectors and economic agents over one year. We compared a modelled closed economy with no rotavirus vaccine with the same closed economy that implemented vaccination to estimate the incremental economic impact across different sectors.

● Our results showed that the modelled economy will reach a surplus following the implementation of a vaccination program. The latter is expected to generate more economic activity with a higher return on taxes and a reduced spending in health care that pays off the total vaccination cost within a given year.

● It is likely that this shift in money flows, measured with this matrix when introducing the rotavirus vaccine may happen rapidly as the vaccine acts on very short term. The model should also be applicable to other frequent vaccine-preventable infectious diseases associated with high indirect costs.

health insurance premiums (B5). Households with children are exposed to yearly epidemics of rotavirus disease where they take days off work for taking care of their sick children. The disease event is expected to reduce their income (A2, B1), spending (B1), tax payment (B3), and the production of the firms (A3). Hence, many different agents will be impacted by the presence of the disease, especially when the disease is frequent and dense during a certain period of the year. Finally, HICs pay healthcare providers for healthcare services delivered to the sick children of their clients, the households (E4).

When introducing rotavirus vaccination, the MoH injects extra money into the system by purchasing vaccines (D6) from manufacturers who, in turn, pay corporate taxes on the profitability of their goods sold (F3). Most importantly, the vaccine is expected to prevent parents being absent from work (A2; B1). Prevention of absenteeism is expected to gen-erate a new balance in the money flow amongst the different agents, with increased productivity, higher wages, more spending, increased tax revenue (A1:B3), and reduced health-care costs (E4). This analytic framework enables us to explore how policy interventions with associated costs will influence changes in the overall money flow in the economy. The out-come measures of interest in a SAM model are two. The first is the difference in total money flow between conditions A and B (G7Avs. G7B,). When G7A< G7B, there is an economic surplus

produced. Subsequently, the net economic surplus can be defined as the value of economic surplus (G7B– G7A) minus

the vaccination cost. The net economic surplus reflects the extra money input introduced by the MoH. The second out-come is the payment offset of the vaccines by the authorities. This outcome measure is obtained through tax revenues and healthcare spending after subtracting condition A from B: [G3 + ABS(G4))>D7, (ABS = Absolute Value].

2.3. Population and epidemiology data

The size of the population in the model is equal to the number of households corresponding to the annual birth cohort in the Netherlands (n = 182,283) [4]. The total annual rotavirus events in children less than 5 years of age in the Netherlands, originated from the literature. Literature data were also used for the tribution of rotavirus-related healthcare visits per rotavirus dis-ease case [4,20–22]. The data used are shown in Table 1. Moreover, a comprehensive list of all variables used in the model are described in Supplemental material 1.

In line with the health economic evaluation of HRV in the Netherlands, for the condition B, reduction rates of the total rota-virus disease equal to 65% where used to quantify the effect of UMV implementation [23]. Vaccine coverage of 90% was assumed. Furthermore, based on the redistribution of case severity following the UMV, a different reduction in resource use of general practi-tioner (GP) visits, hospitalizations and nosocomial infections was applied (88%, 93%, 78%, respectively) with relatively more cases cared at home [23].Table 1also lists at the end the unit cost for each of the medical activities related to rotavirus disease as retrieved from published literature and local reports [23].

2.4. Economic inputs for households and firms

Economic inputs, such as gross earnings of a 25- to 35-year-old individual (in line with the national average age at which young families have children), are distinguished between employed indi-viduals and contract workers. The proportion of the income that was disposable for spending at the individual level originated from the Central Bureau of Statistics in the Netherlands (Table 2) [24,25]. An income tax rate of 40.85% was applied on gross earnings as per gross income tax bracket€20,143 – €33,994 plus a VAT rate of 21%

Figure 1.Square matrix structure of a SAM model.

HIC: health insurance companies; MoF: Ministry of Finance; MoH: Ministry of Health.

on disposable income spent [26,27]. Gross earnings were based on a full-time employment with an annual number of 20 days leave plus 8 public holidays. Annual gross profitability per employee was calculated by dividing the annual profits of firms in the Netherlands by the total workforce in the for-profit sector on which a corporate tax rate of 20% was applied [28,29]. Based on recent research, it was assumed that on average 5.5 working days per family were lost in case they suffered from a rotavirus disease event [5].

2.5. Model assumptions

Out-of-pocket payments for vaccines by households were not considered. Insurance companies were assumed to be

not-for-profit and did not pay corporate taxes. It was assumed that for each rotavirus disease event, employed parents took special leave receiving 70% of their gross income as per the official legislation [30]. Parents who were contract workers were assumed not to receive any income during the workdays lost.

2.6. Sensitivity analysis

Scenario analysis was conducted to assess the conditions resulting in no total money surplus between condition A and B. Another scenario assessed what would happen with condi-tion B, if the MOH selected a specific target group at higher risk for severe disease to be vaccinated, as has been recently proposed in the Netherlands.

Table 1.Input parameters for population, epidemiology, and medical cost of rotavirus disease. Parameter Input value Input (%)

Epidemiology condition A

Annual birth cohort 182,283

Total annual RV events 73,456 40%

Staying home 45,365 25% Medical visit 24,343 13% Hospital 2,940 1.6% Nosocomial 808 0.4% Epidemiology condition B Vaccinated Unvaccinated

Annual birth cohort 164,055 18,228

Total annual RV events 23,139 14% 7,346 40% 65% reduction

Staying home 20,164 12% 4,537 25%

Medical visit 2,629 2% 2,434 13% 88% reduction

Hospital 185 0% 294 1.6% 93% reduction

Nosocomial 160 0% 81 0.4% 78% reduction

Direct health care cost

GP visit €31.8

Hospitalization €2,482

Nosocomial infection €2,253

Vaccine cost €117/course

Reduction rate on tender submission 5%

Vaccine coverage rate 90%

GP: general practitioner; RV: rotavirus

Table 2.Fiscal and other cost input parameters of households, firms, and disease management.

Fiscal input parameter Input value Source

Households

Gross earnings of 25- to 35-year-old m/f €33,700 CBS [24] Gross earnings of an employed 25- to 35-year-old m/f €33,900

Gross earnings of an independent contractor 25- to 35-year-old m/f €32,000 Proportion of employed workers 25- to 35-year-old m/f 88% Proportion of independent contract workers 25- to 35-year-old m/f 12%

Income tax rate 40.85% Belastingdienst [26]

Gross disposable income main breadwinner < 35 years 75% CBS [25]

VAT on G&S 21% OECD [27]

Annual working days 232 Calculated

Annual payment per household for health insurance €1,200 Estimated Firms

Gross profit before taxes €222,097,000,000 National accounts 2016 [28]

Total workforce 6,526,000

Annual profitability per employee €34,033 Calculated

Daily profitability per employee €147 Calculated

Employers contribution when absence for care 70% Rijksoverheid,2018 [30] Lost working days due to rotavirus case 5.5 Standaert et al. 2015 [20] Lost profitability due to RV per employee €815 Calculated

Total lost revenue due to RV €59,842,915 Calculated

Corporate tax rate 20% Belastingdienst, 2018 [29]

CBS: Central Bureau of Statistics; G&S: goods and services; GP: general practitioner; m/f: male/female; OECD: Organization for Economic Co-operation and Development; RV: rotavirus; VAT: value-added tax

Deterministic one-way sensitivity analysis was also conducted to assess the effect of disease epidemiology (frequency and sever-ity proportion), work absenteeism (average days being absent from work), cost of the vaccine, tax payment, proportion of con-tract workers, and tax on profit gain of the firms (Table 3). The selected range per variable was based on literature review or in the absence of data a range of ±20% of the reference value was used. The results of the sensitivity analysis for the selected variables of interest were presented in tornado-diagrams for the two outcome measures of interest: total economic surplus and vaccine cost offsets.

Moreover, a probabilistic Monte-Carlo, sensitivity analysis, using @Risk 7.5.1. was conducted after selecting the most impactful variables of the deterministic one-way sensitivity analysis. The probabilistic analysis assumed that variables fol-lowed various distributions and assessed which combination of conditions reduced or augmented the outcomes observed under the one-way sensitivity analysis.

3. Results

Table 4shows the modeled difference in outcomes between

condition A and condition B. The results of each condition are separately reported in Supplemental material 2.

Economic surplus increased by€45.952 million for one year at the new equilibrium of infection spread after the introduc-tion of the vaccine and the net economic surplus was €26.758 million. The net vaccine cost-offset after paying for the vaccine was€3.366 million. In the presence of UMV, firms were more profitable because of increased labor productivity leading to more corporate tax payment (~€7 million annually). Households with children under 5 years of age would earn an additional €14.359 million of which about €5.836 million was consumed on goods and services (+€32 per year per household). The additional earnings of households and firms were also beneficial to the MoF, which revenues increased by €14.561 million, due to the increased income tax, corporate tax and VAT collected through consumption. The MoH through the HICs saved €7.998 million on averted rotavirus-related healthcare costs. The investment cost for vaccines in UMV, which totaled€19.194 million, was completely offset by the additional income of the MoF and the reduction in spend-ing on healthcare treatment by the MoH (€3,365 million).

Figures 2and3show the results of the deterministic, one-way sensitivity analysis on the outcomes of vaccine cost-offset and of the economic surplus.

The results of the sensitivity analysis show that the economic surplus never falls negative meaning that overall vaccine preven-tion causes economic gains even when the price of the vaccine is at the low end (€94/course) or the frequency of the disease is low (<60,000 events per year). Vaccine coverage has a huge impact on the economic surplus because the surplus is dominated by the vaccine use. The sensitivity analysis for cost-offsets shows that variables determining the cost of treatment like hospital rate, hospital cost, and vaccine effect overall have the largest impact.

The probabilistic sensitivity analysis (PSA) included the fol-lowing variables [Distribution type (Mean; Min; Max)]:

● Total_RV events [Normal (73,456; 49,345; 96,561)]

● Proportion_employed [Normal (0.88; 0.83; 0.91))

● Vaccine-cost [Normal (117; 90; 125)]

● 1-VE overall [Normal (0.35; 0.25; 0.40)]

● Absent when home care (days) [Normal (4; 2; 6)]

Table 3.List of variables tested in the one-way sensitivity analysis.

Variable Mean value Minimum Maximum Absent when home care (days) 4 2.6 5.4 Absent with GP-visit (days) 5 4 6 Absent with hospital visit (days) 9 7.2 10.8 Absent with nosocomial (days) 13 10.4 15.6 Absent when sick own (days) 1.2 0.96 1.44 Proportion employed 88% 84% 92% Total RV events 73,456 58,765 88,147 Hospital cases 2,940 2,352 3,528 Disposable income 75.00% 59.92% 89.88% Corporate tax rate 20% 16% 24%

1-VE overall 35% 28% 42%

Cost hospital €2,482 €1,986 €2,978 Vaccine cost €117 €93.6 €140.4 Vaccine coverage 90% 72% 95% GP: general practitioner; RV: rotavirus; VE: vaccine effect

Table 4.Net transactions between the condition without and with vaccination over one year at equilibrium.

MoF: Ministry of Finance; MoH: Ministry of Health

● Vaccine coverage [Normal (0.9; 0.80; 0.95)]

● Absent with GP-visit [BetaGeneral (5; 4; 6)]

● Cost hospital [BetaGeneral (2,482; 2,247; 2,987)]

● Disposable income [BetaGeneral (0.75; 0.68; 0.8)]

● Corporate tax rate [BetaGeneral (0.20; 0.18; 0.22)]

● Hospital cases [BetaGeneral (2,940; 2,646; 3,400)]

The results of the PSA are shown in Figure 4. The spread of the vaccine cost-offsets is shown on the left side with

the coefficients of the variables that have the biggest impact on the right side. Only a few simulations resulted in a negative outcome through a combination of low total events, high vaccine cost and a low average number of days of being absent from work. Additional data on the PSA results are described in the Supplemental material 3. A SAM analysis focusing on a very specific target group like high-risk children, indicates a negative economic surplus (−1% of the UMV strategy), with a positive balance in vaccine cost-offset

Figure 2.One-way sensitivity analysis on vaccine cost-offset.

Base-case value:€ 3,365,577; light: lower value; dark: higher valueGP: general practitioner; RV: rotavirus

Figure 3.Sensitivity analysis on the economic surplus.

(€885.819 cost gain) (see corresponding matrix in Supplemental material 4).

4. Discussion

The analysis described here assesses the income distribution within a SAM framework and reflects how an infectious dis-ease like rotavirus and its prevention influences the money flow in a broad range of perspectives within a modeled closed economy. Central to the analysis is the way in which reported childhood cases of rotavirus influence employment activities of families. A unique feature of the Dutch labor market is the high proportion of contractors and the legal limitation on sick leave allowances that are taken by an individual in relation to a sick family member, being paid 70% of one’s salary. Under both circumstances it results in a reduced income during the illness period. This loss in income reduces the money flow across different sectors of the economy. The MoF will earn less in direct income taxes and indirect taxes from consumption. Similarly, the money flow of less disposable income for families, reduces the demand for goods that may influence the profitability of firms. In addition, averted rotavirus disease events may reduce healthcare spending. As these events often occur during winter periods where young children are most exposed to many different infectious diseases, they cause bottleneck conditions in healthcare delivery of emergency rooms and hospital pediatric wards. Introducing vaccination against pediatric infectious diseases therefore reduces pres-sure on hospital resources and improves the overall quality of care [5,31]. The cost savings may be used to accomplish a decrease in insurance premiums for households, since HICs are not for profit in the Netherlands.

Resource-allocation decisions are often isolated in evaluating healthcare costs associated with medical conditions. The SAM framework highlights the weakness of applying a health sector framework only to decision making as it ignores the money flows in the economy influenced by illness and how changes in health status impact upon them. It is possible to reject this notion suggesting that lost working hours can be made up, and

other workers can provide additional labor during absences, however, the wage losses of these workers represent real income losses. While output can be reallocated to other work-ers, all things being equal, more output and efficiency are likely to be achieved by avoiding health events and chronic condi-tions, which is the premise behind the extensive literature linking health to labor productivity and economic growth [32– 34]. Within our framework these losses can be accounted for due to reduced annual income, although small, when magni-fied across the economy, will represent reduced money flows. The perspective of the SAM framework is likely intuitive to employers who understand the costs and losses associated with paying for sick leave with reduced output. Additionally, these considerations are important with respect to the numbers of independent contractors that are on the rise globally.

For decades, the SAM framework has been applied, originally in developing economies, to assess poverty reduction following a capital injection. It has been more recently applied to mea-sure the effect of poor health where a United Kingdom (UK) study projected the economic impact of pandemic influenza [35]. It was shown that school closure due to pandemic influ-enza but also large prophylactic absence from work were asso-ciated with considerable loss of gross domestic product (GDP). Another analysis using a SAM framework concluded that it was not the actual cost of antimicrobial resistance but the possibility of future costs in terms of world-wide GDP losses that reflected the current disease burden and the need to address antimicro-bial resistance [36]. This was in line with previous findings in the UK where antimicrobial resistance was estimated to amount to a GDP fall between 0.4%-1.6%, which is equivalent to a £3 –-11 billion loss in monetary terms [37].

The prototype analysis presented here reflects multiple perspectives and the impact of rotavirus cases on households, government (MoF & MoH), industry and health services within a 1-year timeframe. Variation in any parameter of the model may influence the outcome. SAM frameworks therefore can also help indicating decision-makers who should pay or co-pay for the vaccine: who is benefiting most, the household, the firms, or the authorities and what may happen if the

Figure 4.Displaying the PSA results of the vaccine cost-offset simulation.

Left side: spread of the outcome measure. Right side: coefficients of influence on the spreadGP: general practitioner; PSA: probabilistic sensitivity analysis; RV: rotavirus; VE: vaccine effectiveness

payment is spread across different agents? Should authorities incentivize the firms for paying for the vaccines linked to a tax reduction of their profit increase because of lower treatment payment in healthcare by the MoH?

For instance, changes in annual rotavirus disease events that occur will influence the money flows across different economic sectors considered in the SAM as illustrated in the sensitivity analysis [38]. However, as this framework represents multiple perspectives, any change in one parameter is likely to generate positive and negative economic changes in the other perspectives. Additionally, at present there are no objective functions for SAM frameworks in health and healthcare with accepted decision rules. But the aim of this work is to intro-duce this new approach for evaluating national vaccine invest-ments on which to build value assessinvest-ments in health and healthcare that move beyond the conventional evaluations. One point we like to make however is that it is difficult to achieve low or negative economic surplus results, keeping in mind that we go for UMV-design. The negative condition can be obtained if the focus is on small target groups with a high treatment cost avoided and a low cost for the vaccine because only a small group needs to be reached. The question raises how easily this specific group can be reached and covered by the vaccine. Also interesting to observe is that the limitation in higher vaccine costs that could increase the economic surplus, will be limited by the vaccine cost-offset that will switch to negative values if the price of the vaccine is too high, com-bined with a low frequency of the disease, low treatment cost, and a limited duration in absenteeism.

As with every modeling exercise, the SAM framework that focuses on money flow distribution in the economy related to rotavirus disease in children has limitations. First, the approach only reflects a subcategory of the national economy. It works with a disaggregated sub-matrix of a national SAM for the Netherlands. It is likely that there will be money flows from stakeholders within the closed economy to stakeholders outside and vice versa. Hence, the total receipts (income) and expenditures of each sta-keholder within this modeled closed economy are not equal [39]. For instance, the sum of the wages paid by the firms to the households, is not equal to the total sum of the money spent by the households (second column). The reason for that is that we excluded from the analysis money at the level of a household that is not spent, and which would lead to tax revenue for the authorities. Second, the average profitability per employee has been calculated based on the average total profitability of com-panies divided by the total workforce. This might not be an accurate reflection of a firm’s profitability per employee. In addi-tion, the vaccine price reflects the current list price but does not include rebates offered during the tender process. Another limita-tion of the SAM framework is its dependence on the dynamic labor market. Changes in average wages or policy changes in labor law and regulations will inevitably influence outcomes of the SAM framework, hence the 1-year time horizon. Due to country-specific input data, the present model outcomes are likely to be most applicable to the Netherlands and transferability to other econo-mies needs to be explored and adjusted.

The framework described here likely underestimates some of the benefits achieved through introducing rotavirus

immunization. For example, the benefits of a herd effect which would reduce the overall amount of rotavirus in the community during the uptake period of the vaccine is not captured within the one-year time horizon measured at equilibrium [40]. On the other hand, we know that the vaccine may cause a risk for an adverse event called intussusception that needs specific medical care. The frequency of the event is low (+6/100,000 children) and the cost for the medical care has not been considered in the current model [41]. Additionally, although uncommon, every year, there are a limited number of deaths attributed to rotavirus that are likely to be avoided through immunization. Again, the short timeline applied here would not capture the lifetime indirect benefits attributed to immunization. Furthermore, as described above, rotavirus immunization has been shown to improve the quality of care in hospitals due to reduced demands on hospitals during seasonal outbreaks. This would likely translate into improvements in care that would improve collateral outcomes and likely save additional resources not accounted for in this analysis.

5. Conclusion

Conventional economic evaluations of healthcare interven-tions often do not assess a range of perspectives and how a disease influences different sectors of the economy and the overall economy. The present study shows that the SAM model could be used to reflect all the consequences of a disease burden on the money flows in different sectors of the economy and how likely that is impacted through new interventions like prevention due to a new health policy.

Acknowledgments

The authors would like to thank Business & Decision Life Sciences platform for editorial assistance and manuscript coordination, on behalf of GSK. Amandine Radziejwoski coordinated manuscript development. The authors retained the final editorial control over all published material.

Author contributions

All authors participated in the design, analysis and interpretation of the study as well as in the development of this manuscript. All authors had full access to the data and gave final approval before submission.

Funding

GlaxoSmithKline Biologicals S.A. funded this study (GSK study identifier: HO-18-19703) and all costs related to the development of the publica-tions. The sponsor had a role in article preparation, study design, data analysis and review of drafts.

Declaration of interest

B Standaert is an employee of the GSK group of companies and holds stock in the GSK group of companies. G Haitsma, MP Connolly and N Kotsopoulos received funding for the analysis and manuscript development by the GSK group of companies. G Haitsma, MP Connolly, and N Kotsopoulos hold no financial interests in the sponsor organization. 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 materials discussed in the manuscript apart from those disclosed.

Reviewer Disclosures

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

Trademark statement

Rotarix is a trademark of the GSK group of companies.

ORCID

Mark P. Connolly http://orcid.org/0000-0002-1886-745X

Baudouin Standaert http://orcid.org/0000-0001-6801-9654

References

Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers.

1. Hasso-Agopsowicz M, Ladva CN, Lopman B, et al. Global review of the age distribution of rotavirus disease in children aged <5 years before the introduction of rotavirus vaccination. Clin Infect Dis.

2019Jan 28;69(6):1071–1078. PubMed PMID: 30689799.

2. Poelaert D, Pereira P, Gardner R, et al. A review of recommenda-tions for rotavirus vaccination in Europe: arguments for change. Vaccine.2018Apr 19;36(17):2243–2253. PubMed PMID: 29576308. 3. Bruijning-Verhagen P, Mangen M-J-J, Felderhof M, et al. Targeted

rotavirus vaccination of high-risk infants; a low cost and highly cost-effective alternative to universal vaccination. BMC Med.

2013;11(1):112. PubMed PMID: 23622110.

• Gives an overview of the epidemiology and the current man-agement of rotavirus disease in the Netherlands including an approach on vaccinating target groups.

4. Goossens LM, Standaert B, Hartwig N, et al. The cost-utility of rotavirus vaccination with Rotarix (RIX4414) in the Netherlands. Vaccine.2008Feb 20;26(8):1118–1127. PubMed PMID: 18215445. 5. Scarbrough Lefebvre CD, Terlinden A, Standaert B. Dissecting the

indirect effects caused by vaccines into the basic elements. Hum Vaccin Immunother.2015;11(9):2142–2157. PubMed PMID: 26186100; PubMed Central PMCID: PMCPMC4635729.

• First economic evalution of rotavirus vaccination in the Netherlands

6. Karafillakis E, Hassounah S, Atchison C. Effectiveness and impact of rotavirus vaccines in Europe, 2006–2014. Vaccine. 2015;33 (18):2097–2107. PubMed PMID: 25795258.

7. Rheingans R, Amaya M, Anderson J, et al. Systematic review of the economic value of diarrheal vaccines. Hum Vaccin Immunother.

2014;10(6):1582–1594. PubMed PMID: 24861846.

8. World Health Organization [Internet]. WHO guide to identifying the economic consequences of disease and injury; 2009. [cited 2019 Oct 3]. Available from:https://www.who.int/choice/publications/d_ economic_impact_guide.pdf

9. Standaert B, Harlin O, Desselberger U. The financial burden of rotavirus disease in four countries of the European Union. Pediatr Infect Dis J.2008;27:S20–7.

• A good evaluation of the financial disease burden of rotavirus disease in the Netherlands

10. Black DC. Working for a healthier tomorrow. London: TSO;2008. cited 2018 Nov 19. Available from:https://www.rnib.org.uk/sites/ default/files/Working_for_a_healthier_tomorrow.pdf

11. Schofield D, Shrestha R, Percival R, et al. The personal and national costs of CVD: impacts on income, taxes, government support pay-ments and GDP due to lost labour force participation. Int J Cardiol.

2013;166(1):68–71. PubMed PMID: 22018513.

12. Connolly MP, Kotsopoulos N, Postma MJ, et al. The fiscal consequences attributed to changes in morbidity and mortality linked to investments

in health care: a government perspective analytic framework. Value Health.2017;20(2):273–277. PubMed PMID: 28237208.

• Shows the importance of fiscal evaluation in health care 13. Kotsopoulos N, Connolly MP. Is the gap between micro-and

macro-economic assessments in health care well understood? The case of vaccination and potential remedies. J Mark Access Health Policy.

2014;2(1):23897. PubMed PMID: 27226842.

14. Keogh-Brown MR, Wren-Lewis S, Edmunds WJ, et al. The possible macroeconomic impact on the UK of an influenza pandemic. Health Econ.2010;19(11):1345–1360. PubMed PMID: 19816886. 15. Keogh-Brown MR, Smith RD, Edmunds JW, et al. The

macroeco-nomic impact of pandemic influenza: estimates from models of the United Kingdom, France, Belgium and The Netherlands. Eur J Health Econ.2010;11(6):543–554. PubMed PMID: 19997956. 16. Imai C, Hall L, Lambert SB, et al. Presenteeism among health care

workers with laboratory-confirmed influenza infection: A retrospective cohort study in Queensland, Australia. Am J Infect Control. 2019. PubMed PMID: 31515100. DOI:10.1016/j.ajic.2019.07.024.

17. Leadership group on Social Accounting Matrices. Handbook on social accounting matrices and labour accounts. Luxembourg: Eurostat Working Papers;2003. 3/2003/E/23.

•• Reference document on how SAM models should be developed

18. World Bank [Internet]. Social accounting matrices and SAM-based multiplier analysis. [cited2018Nov 16]. Available from:http://siter e s o u r c e s . w o r l d b a n k . o r g / I N T P S I A / R e s o u r c e s / 4 9 0 0 2 3 -1121114603600/14017_chapter14.pdf

19. Kotsopoulos N, Connolly MP, Postma MJ, et al. Fiscal consequences of changes in morbidity and mortality attributed to rotavirus immunisation. Vaccine.2013;31(46):5430–5434.

20. Standaert B, Van de Mieroop E, Nelen V. Exploring the potential impact of rotavirus vaccination on work absenteeism among female administrative personnel of the City of Antwerp through a retrospective database analysis. BMJ Open. 2015;5(6):e007453. PubMed PMID: 26129633; PubMed Central PMCID: PMCPMC4486949. • An interesting way how to precisely measure the work

absen-teeism pro- and after the rotavirus vaccine introduction 21. Standaert B. Are changes occurring in the perceived value of

vaccines? Belgian J Pediatr.2016;18(4):325–327.

22. Tran AN, Husberg M, Bennet R, et al. Impact on affected families and society of severe rotavirus infections in Swedish children assessed in a prospective cohort study. Infect Dis (Lond).2018;50 (5):361–371. PubMed PMID: 29260605.

23. ZorgInstituut Nederland. Farmaco-Economisch rapport voor humaan rotavirus, monovalent (Rotarix®) als actieve immunisatie, bij zuigelingen van 6 tot 24 weken, ter voorkoming van gastro-enteritis veroorzaakt door een rotavirusinfectie. Diemen: Zorginstituut Nederland.2017Jul 20.

24. Central Bureau of Statistics [Internet]. Werkzame beroepsbevolking; gemiddeld inkomen; 2018 [cited 2018 Aug 16]. Available from:

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83686NED/table? ts=1519746440001

25. Central Bureau of Statistics [Internet]. Inkomensverdeling van huishoudens; nr, 2005-2014;2018. [cited 2018 Aug 16]. Available from:https://opendata.cbs.nl/statline/#/CBS/nl/dataset/82959NED/ table?ts=1532525633247

26. Belastingdienst [Internet]. Inkomstenbelasting; Heffingskortingen, boxen en tarieven; Boxen en tarieven; Overzicht tarieven en schijven;2018. [cited 2018 Aug 16]. Available from:https://www. belastingdienst.nl/wps/wcm/connect/bldcontentnl/belastingdienst/ prive/inkomstenbelasting/heffingskortingen_boxen_tarieven/ boxen_en_tarieven/overzicht_tarieven_en_schijven/u-hebt-in-2018-de-aow-leeftijd-nog-niet-bereikt

27. OECD. Consumption Tax Trends 2016 -: VAT/GST and excise rates, trends and policy issues. Paris: OECD Publishing;2016, DOI:10.1787/ ctt-2016-en.

28. Central Bureau of Statistics. National accounts of the Netherlands. The Hague (NL): Statistics Netherlands;2016.

29. Belastingdienst [Internet]. Winst, vennootschapsbelasting, Tarieven vennootschapsbelasting;2018[cited 2018 Aug 16]. Available from:

https://www.belastingdienst.nl/wps/wcm/connect/bldcontentnl/ belastingdienst/zakelijk/winst/vennootschapsbelasting/tarieven_ vennootschapsbelasting

30. Rijksoverheid [Internet]. Rijksoverheid, home, Onderwerpen, Zorgverlof, documenten. [cited August 16,2018]. Available from:

https://www.rijksoverheid.nl/onderwerpen/zorgverlof/documenten

31. Dort T, Schecroun N, Standaert B. Improving the hospital quality of care during winter periods by optimizing budget allocation between rotavirus vaccination and bed expansion. Appl Health Econ Health Policy.2018;16(1):123–132. PubMed PMID: 29159785; PubMed Central PMCID: PMCPMC5797246.

32. Suhrcke M, McKee M, Arce RS, et al. The contribution of health to the economy in the European Union. Luxembourg: Office for Official Publications of the European Communities;2005. 33. Van Zon A, Muysken J. Health as principal determinant of

eco-nomic growth. In: López-Casasnovas G, Rivera B, Luis Currais L, editors. Health and economic growth: findings and policy implica-tions. Cambridge (MA): MIT Press;2005. p. 41–63.

34. Bloom DE, Canning D, Sevilla J. The effect of health on economic growth: theory and evidence. National Bureau of economic research working paper series; 2001. (NBER Working Paper no. 8587).

35. Smith RD, Keogh-Brown MR, Barnett T, et al. The economy-wide impact of pandemic influenza on the UK: a computable general equilibrium modelling experiment. BMJ.2009;339:b4571. 36. Taylor J, Hafner M, Yerushami E, et al. Estimating the economic

costs of antimicrobial resistance: model and Results. Santa Monica (CA) and Cambridge (UK): RAND Corporation;2014.

37. Smith RD, Yago M, Millar M, et al. Assessing the macroeconomic impact of a healthcare problem: the application of computable general equilibrium analysis to antimicrobial resistance. J Health Econ.2005;24(6):1055–1075.

38. Verberk JDM, Pijnacker R, Bruijning-Verhagen P, et al. Biennial pattern of rotavirus gastroenteritis in The Netherlands and a shifting age distribution after a low rotavirus season, 2010–2016. Pediatr Infect Dis J. 2018;37(9):e248–50. PubMed PMID: 00006454-201809000-00027.

39. Stuttard N, Frogner M. Developing a pilot social accounting matrix for the United Kingdom. Econ Trends.2003;594:80–93.

40. Pollard SL, Malpica-Llanos T, Friberg IK, et al. Estimating the herd immunity effect of rotavirus vaccine. Vaccine. 2015. 33 (32):3795–3800. PubMed PMID: 26116250.

41. EMA: summary of product characteristics [Internet]. 2017. [cited 2018 Jan 8]. Available from:http://www.ema.europa.eu/docs/en_ GB/document_library/EPAR_-_Product_Information/human/ 000639/WC500054789.pdf