Seasonal influenza in children: Costs for the health system and society in Europe

Abstract Background Pediatric influenza causes significant morbidity annually, resulting in an increased economic burden. Therefore, we aimed to summarize existing literature regarding the costs of pediatric influenza in Europe, paying particular attention to the direct and indirect costs considered in the economic evaluations. Knowing health and social costs of childhood influenza is essential to support value‐based health decisions to implement effective immunization strategies. Methods We searched three databases for articles published to September 3, 2021. Eligible studies were those reporting the economic burden of influenza in the pediatric and youth population in European countries written in English language. Results Overall, 2225 records were screened, and 9 articles were included. Costs estimates are different across countries and in the age groups considered. Direct costs per episode, whose major expense driver are hospitalizations and pediatric examinations, range from about €74 in Italy to €252 in Germany. Important variations are observed based on age, with the youngest group absorbing in some cases double the resources of the older ones such as (in Italy, in France and in Germany). Regarding indirect costs, workdays lost by parents resulted in higher costs for children <2 years and 2–5 years than those >5 years of age and their economic impact was variable reaching €251 per week in Germany. Conclusion Evidence obtained in our review strengthened the awareness about the economic impact, in terms of direct and indirect costs, of pediatric influenza requiring, as a priority action in Europe, the implementation of influenza vaccination policies in this target population.


| INTRODUCTION
Seasonal influenza causes significant morbidity and mortality annually, especially in vulnerable individuals such as children and the elderly, resulting in an increased economic burden on both health system and society. 1 Indeed, every year, seasonal influenza epidemics result in about 1 billion cases, of which 3-5 million are severe cases, especially among the vulnerable groups, and about 250,000 to 500,000 deaths worldwide. 2 In Europe, influenza causes 4-50 million symptomatic cases annually, about 15,000-70,000 3 deaths and 150,000 influenza-related 4 hospital admissions. Specific vulnerable groups (the elderly, patients with chronic diseases and comorbidities, younger children [<5 years of age], and pregnant women) are at increased risk of developing severe illness, complications and dying from influenza. 5 Anyone can get influenza, but infection rates are highest among the pediatric population ($20%-30% annually), 6,7 which presents a great risk of being infected due to limited pre-existing immunity of children. 8 Additionally, several studies have shown that young children play an important role in transmitting the influenza to their families and the wider community. [9][10][11] The literature suggests that influenza vaccination of young children may reduce disease rates in non-immunized individuals in the local community. [12][13][14] Furthermore, although most influenzarelated complications and deaths occur among the elderly and people with underlying chronic conditions, 15 children younger than 5 years of age are at high risk of developing serious influenza-associated disease, with indirect impact on their siblings and parents. 8 Actually, worldwide, each year approximately 870,000 children under 5 years of age have a hospitalization attributable to influenza, and it is estimated that between 28,000 and 111,500 deaths in this age group are attributable to influenza-related causes, the vast majority of which occur in developing countries. 16 Furthermore, it is estimated that the average annual rate of outpatient visits attributable to influenza are approximately 10, 100, and 250 times as high as hospitalization rates for children 0-5 months, 6-23 months, and 24-59 months of age, respectively. 17 Among children, influenza has been considered serious only in those with chronic diseases and with medical conditions at higher risk of developing complications but in recent years the burden in healthy children has become increasingly explicit also thanks to the scientific evidence produced in the field of socio-economic studies. Indeed, childhood seasonal influenza can impose substantial socio-economic burden on healthcare system, families, and society in terms of hospitalizations, outpatient visits, medications, school absence, and missed workdays, either due to secondary illness in a caregiver or to the need to care for a sick child. 18 An accurate assessment of the economic burden of influenza at the population level could guide policy decisions regarding influenza vaccination in children that is still widely debated in the European countries. 4 As such, the aim of this systematic review is to summarize existing literature regarding the economic burden of childhood influenza per episode in Europe, paying particular attention to the type of costs (direct costs such as medical examinations, hospitalizations, drugs prescription, and indirect costs such as workdays lost by parents and children school leave) considered for the evaluation of the economic burden. In fact, investigating and knowing the healthcare and social costs of influenza in children is essential to support value-based health decisions and to implement effective immunization strategies among children and youths in European Union (EU).

| Search strategy
A systematic review of the academic literature was conducted and reported according to the Preferred Reporting Items for Systematic Reviews (PRISMA). 19 The literature search was performed by consulting three databases, namely, PubMed, Web of Science (WoS) and Scopus. The following search string was used on PubMed: "(("influ- The articles records were entered in an Excel work sheet and screened according to the inclusion/exclusion criteria. A check for duplicates was performed; the selection was made firstly by reading titles and abstracts, and then the full texts.

| Inclusion/exclusion criteria
All studies focused on the economic burden of influenza in the pediatric and youth population (age range 0-18 years) were considered potentially eligible. We included original articles and systematic reviews exclusively in English language and conducted in European Countries. No time limits were placed on the search, thus including all papers up to September 3, 2021. Studies describing the economic impact of influenza in populations aged >18 years and those concerning non-European contexts were excluded. Narrative reviews, commentary, editorials, conference presentation, and citations not provided with full text were excluded as well as studies conducted in animals or in vitro.

| Selection process and data extraction
Two researchers (L. V. and F. D'A.) independently screened titles and abstracts first and full texts afterwards. Any disagreement was resolved by discussion or by the involvement of a senior researcher (G. E. C.).
From the articles definitively included in the literature review, the following information were extracted: first author's name, publication year, country, study design, target population, setting, study duration and time, and main findings related to the economic burden of influenza, considering both direct and indirect costs.
When included, the systematic reviews were subjected to the snowballing process, using the bibliographic references and citations in the reviews in order to identify additional articles that met the inclusion criteria of this review.

| RESULTS
The database search, after duplicates removal, brought a total of 2225 records. After an initial selection by titles and abstracts, 64 full-text articles were selected. Following the inclusion and exclusion criteria, the screening resulted in the final inclusion of nine articles. Details about the study selection process are shown in the flowchart ( Figure 1). Of the nine studies included in our systematic review, four (44.5%) were retrospective studies, 20-23 one was a case-control study (11.1%), 24 two were prospective studies (22.2%), 25,26 and two (22.2%) systematic reviews. 8,27 No new studies were included after the snowballing process, because the two systematic review included studies that were already selected. The two systematic reviews reported international data and European information for the following countries: Italy, Germany, France, the Netherlands, Austria, Finland, Spain, and the United Kingdom. 8,27 The first systematic review published in 2012 8 summarized influenza burden-especially in terms of health outcomes but also considering its economic impact-in children in Western Europe, emphasizing the significant direct impact of influenza on sick children and its indirect impact on their siblings and parents. Instead, the second revision of the 2018 27 examined studies on the costs of Influenza-Like Illness (ILI) in high-income countries and considering all age groups of the population. Tables 1 and 2 report the details of each study, excluding the two systematic reviews. Among the seven primary studies included in our study, two (29%) were conducted in Germany, 22,24 two (29%) in Italy, 20,25 one (14%) in Sweden, 23 one (14%) in France, 26 and one (14%) in Belgium. 21 Ehlken    The study conducted by Crott et al. 21  The costs associated with influenza also vary in relation to the influenza virus type. In fact, according to Esposito et al., 25 influenza A cases were significantly more expensive than influenza B cases and this could be due to the fact that influenza A virus appears to cause more severe diseases than influenza B virus. 25,28 Eventually, influenza in children aged <2 and 2-5 years was significantly more expensive than in children aged >5 years. The difference was mainly associated with the indirect costs due to parents' lost working days and could be justified by the evidence suggesting that influenza A occurs more frequently among children <4 years of age than in older children. 28,29 Therefore, it is evident that influenza in children aged <5 years is more expensive than in older children. 25 In this age group, the higher costs are linked both to an increase in direct costs for hospitalizations, medical visits and medications, and, above all, to indirect costs linked to the missed workdays, either due to secondary illness in caregivers, namely, parents, or their need to care for a sick children. 21,22,[24][25][26] As regards direct costs, those related to hospitalizations have an important impact and are higher for children aged <1 years than in older. 24 Hospitalization costs increase in the case of flu complications especially in younger children. Scholz et al. 24 showed higher costs in children with pneumonia and aged <1 years. Also Ehlken et al., 22 showed that total direct costs per influenza/ILI episode with at least one com- to be as high as 30% in the United States. 31 Not only these practices prompt unnecessary costs to the healthcare system but they may also favor the emergency of drug-resistant infections in the long term. 27 Considering this type of information in pharmacoeconomic studies could provide useful evidence for decision and policymakers.
In different studies included in our systematic review, indirect costs were shown to contribute most to the average total cost of influenza. [23][24][25] In particular, workdays lost by parents resulted in higher indirect costs for children <2 years and 2-5 years than those >5 years of age and are higher for workdays lost by mothers. 25 Furthermore, the role of children in influenza transmission to their elderly contacts was demonstrated by Ghendon et al. 32   To achieve this goal, there is the need for aligning and sharing of international guidelines, taking into account the broad value of vac- However, the overview that we have provided pinpointed several meaningful aspects, namely age-dependent differences in costs, contribution of indirect costs to the total cost, which can help the appraisal of the value of influenza vaccination in decision-making. writing; validation; supervision.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12991.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.