Rotavirus disease is associated with a substantial financial burden. The virus is a major cause of acute gastroenteritis (AGE) in infants and children under the age of 5 years worldwide.1–4 Each year in industrialized countries, rotavirus AGE is responsible for an estimated 223,000 hospitalizations, 1.8 million outpatient visits, and 7.1 million episodes of home care.2 Thus, rotavirus AGE results in considerable medical resource utilization and substantial costs to national health care payers, families of patients and employers.5 In Europe, these costs have only been studied in a limited way and in a few countries.6
The total cost evaluation of a disease has a significant added value if positioned in the right context. This context is normally the evaluation of the total disease burden which, in addition to costs, includes the clinical consequences at population level (epidemiology) and the Quality of Life (QoL) impact. This complete set of information is essential for the economic evaluation of new treatment options emerging in the market and as such it may be requested from health care authorities as part of the assessment for their policies.
The impact of new interventions on clinical outcomes and on QoL is most often measured and reported through randomized clinical trials. Total cost impact in contrast is generally more complex to assess and may be estimated using modeling techniques. This presents some specific challenges; one is to include the appropriate cost items in the analysis to reflect the cost perspective under consideration such as that of the patient, the health care provider, the third party payer, or the society. Furthermore, it is impossible to report an overall cost across different countries because cost burden is country-specific. For instance, a treatment resource may be used more often if it is relatively inexpensive, therefore the treatment uptake will influence the disease outcome and total management cost of the disease. Price differences between countries may therefore influence the total cost picture.
In terms of managing diarrhea in infants and children caused by rotavirus infection, resource use and cost per case are well documented because the disease itself and the different treatment options are well-defined.1,3 The remaining unknowns are the exact frequency of the disease per year, its distribution across various age groups and the proportion of the population following the different treatment patterns available.
There are several methods for capturing epidemiological and financial information relating to a disease. The most accurate is the application of prospective, observational cohort studies with a duration of at least 1 year. An alternative method is the modeling approach that mimics the country-specific disease distribution per year plus the country-specific treatment options. Modeling allows the prediction of values missing in real life, such as the total number of diarrhea events and associated emergency visits.7,8
In the present study, we used the modeling approach to assess the financial burden of rotavirus disease in 4 countries of the European Union: Belgium, France, the Netherlands, and the United Kingdom. We compared the cost of rotavirus disease in terms of direct medical and indirect costs and explored new ways of reporting the results.
A decision tree model has been selected to investigate the financial burden caused by rotavirus disease in European countries. We selected countries for which sufficient reliable background data are available or easily accessible: Belgium, France, the Netherlands, and the United Kingdom. This selection enabled us to make a comparison between 2 relatively small and 2 larger European Union countries.
We constructed a Markov cohort model9 developed in TreeAge software (http://www.treeage.com/) (Fig. 1). This model reflects the change in disease occurrence adjusted by subject age. The model also includes the different treatment patterns for rotavirus disease, and can be adjusted to the pattern specific to each country. For example, the Netherlands and the United Kingdom have an organized medical telephone service where the caller can obtain treatment advice; this is a form of paid support which is not available in other countries. The model is calibrated to the overall disease frequency estimates by country per year over the different age groups and time periods.
Model assumptions specific to the disease or country include the following:
- From the underlying disease pattern for rotaviral disease,9–11 a Weibull distribution12 of probability of infection over the first 70 months of life was constructed with the following parameters: shape coefficient 1.5 and scale coefficient 24.2 (Fig. 2). These parameter values were confirmed by a multicenter prospective study of the burden of rotavirus acute gastroenteritis in Europe, the REVEAL study.13
- Overall frequency data on rotavirus diarrhea events in children under 5 years of age are absent. Only data on children seeking medical advice are available. Unless specified, we assumed conservatively that overall a minimum of 40% of the subjects of one birth cohort would suffer from rotavirus diarrhea before 5 years of age and that medical advice would be sought for a maximum of 1 in 2 sick children. The values are comparative estimates based upon studies conducted in France9 and were submitted to sensitivity analysis (see below).
- Full breast-feeding confers protection against viral diarrhea events.14,15 We assumed that at least 50% of infants are breast-fed at birth with an exponential decrease thereafter (β-scale coefficient = 2).
- Only severely ill children for whom medical advice is sought will be sent to hospital. The severity scale used here is based on a Vesikari score of 11 or more out of 20 points.16
- Nosocomial infections occur in a maximum of one-third of young children hospitalized for causes other than community acquired rotavirus diarrhea, and they are age-dependent.17
Two cost perspectives were considered in this analysis. One is the authority in a country funding the medical costs; the other is the society that includes all the other costs related directly or indirectly to the management of the disease under study. Unit cost data by country for direct medical costs reimbursed or paid by the health authorities in a country were collected from national databases. However, one should be aware that country-specific reimbursement costs vary slightly within short time periods and one should therefore check for regular updates of these cost values. Table 1 presents the unit costs for medical (first-line), emergency, and hospital visits, plus the unit costs for treatment. To perform cross-country comparisons, we adjusted the unit cost values with the Purchasing Power Parity (PPP) health care exchange rates per country provided by the Organisation for Economic Cooperation and Development (OECD).18 No discount rate was applied on the cost figures as the analysis reports costs per child per year.
Indirect costs were estimated by considering the loss of productivity of the parents of children with rotavirus diarrhea using the human capital approach.31 The data for out-of-pocket costs such as additional nappies, copayment for drugs and medical visits, transport, and parking costs are not readily available or are difficult to assess at the country level, therefore they were not included in this analysis.
To obtain an accurate assessment of lost productivity at country level, we analyzed the legislation of social security regarding maternity, parental and sick child-related leave for paid jobs in each country (Table 2). The period of legislated, paid maternity leave was included in the analysis. This period does not allow for accounting an indirect cost during maternity leave. Because of the lack of detailed data on parental and sick child-related leave these issues were too complex to be considered and accounted for in this model. Therefore, we estimated a minimum and maximum value of indirect cost due to the impact of rotavirus diarrhea for each country in the following way: the model generates the number of days of sick child-related leave (as a result of the child's diarrhea) for 1 year (post-maternity leave). This value was multiplied by the reported proportion of employed women in the age-range of 15-39 years and the average payment per working hour per women per country. Data on women in the workforce were obtained from the official European statistics database for each country (Table 3). These calculations are considered to yield the maximum estimate of indirect cost by country. The minimum estimate assumed that only half the women in the workforce have a paid job during the first and subsequent years postpartum. Many mothers choose part-time work during that period or benefit from parental leave as authorized in their country. The true value of the indirect cost estimate should fall between the estimates of maximum and minimum values. Lastly, we investigated which group (authorities, employers, and/or employees) paid the most in terms of indirect costs by country.
Multiple probabilistic sensitivity analysis was performed on direct medical costs with TreeAge software on 2 aspects of the input data using distribution estimates: proportion of children with rotavirus AGE and unit cost data (Table 4).
The overall results are reported as absolute costs in Euros (€), per child under 5 years of age and per country. In the multiple probabilistic sensitivity analysis, the model was run in second-order Monte-Carlo simulation12 with 1000 iterations for each country and reports a cost distribution per country per child under 5 years of age.
The observed values and modeled annual estimates of the total number of rotavirus-related AGE events in children under the age of 5 years in each country are shown in Table 5. The model accurately reproduced the known values such as the number of medical visits and the hospitalizations for each country. In addition, the model generated estimates for some variables such as rotavirus diarrhea events and severe rotavirus diarrhea events that are not available at country level.
Numerical values assembled in Table 5 were then used to estimate the total direct medical costs per year and the direct cost per child under the age of 5 years (Table 6). These costs are presented unadjusted and PPP-adjusted and are tabulated as major cost items for each country per year. The unadjusted cost per child per year varied from €7.97 to €13.64, and the PPP-adjusted cost from €7.67 to €12.26.
Table 7 shows the estimated PPP-adjusted indirect costs with minimum and maximum values, overall and per child per year. The range of the indirect cost per child per year is estimated at €10.22–€20.44 in the Netherlands, €7.54–€15.07 in France, €6.83–€13.65 in Belgium, and €10.31–€20.63 in the United Kingdom.
The calculation of the grand total of direct medical and indirect costs related to rotavirus AGE is reported in Table 8. The sums of both, the direct medical and the indirect costs, reveal that the calculated cost per child of the annual birth cohort and per country is around €23.11 ± 0.70 (±3%). In other words, all 4 countries spend very similar amounts per child per year for the total management of rotavirus AGE.
Figure 3 shows the results from the multiple probabilistic sensitivity analysis of PPP-adjusted direct medical costs per child, per country, and per year, presenting the cost range over which the direct medical cost might vary in each country. Countries spending more money in direct medical costs have a higher average value with a higher standard deviation or a wider spread in their costs figures. The average values in the figure deviate slightly from the reported cost figures in Table 6 because the results in the figure are skewed after Monte-Carlo simulation.
Contrary to what one might expect, estimating the total management cost of a disease at country level is not a straightforward exercise. Although the epidemiology of rotavirus disease, its distribution as a function of age and its annual peak during the winter period may be similar across the different countries, its management and its related costs vary considerably. The use of available healthcare resources depends upon the specific structure of each country's healthcare system. In the United Kingdom and the Netherlands, more emphasis is placed on first-line intervention, limiting the use of the more costly second-line healthcare support systems such as emergency services and hospitals. By contrast, Belgium and France provide open access to second-line interventions sooner during the disease process, incurring a higher average direct cost per child for the treatment of rotavirus disease. Consequently, the average direct medical costs per child per year vary by an approximate cost difference of €4.6 between the most and the least expensive country.
The probabilistic sensitivity analysis suggests that healthcare systems such as those in the Netherlands and the United Kingdom may manage rotavirus disease more efficiently compared with the systems in France and Belgium as they do not seem to have comparatively more diarrhea cases or specific deaths, yet their average global medical management costs per child are lower. This is also reflected in the curves of Figure 3 where France and Belgium have much larger standard deviations around average costs than the Netherlands and the United Kingdom.
In all 4 countries, 70–80% of the total direct medical costs are because of hospitalization including community acquired and nosocomial infections. The indirect costs are, however, highest in the Netherlands and the United Kingdom, contrary to what is observed for the direct medical costs. The surprising result is that the total costs of AGE are very similar in all 4 countries studied, with the difference between the most and the least expensive country being marginal (€1.53). This result demonstrates that a direct medical cost analysis alone would only have revealed part of the total relevant costs. One can conclude that the more emphasis is placed on first-line treatment and parental care of sick children, the higher the indirect costs are for society, as the comparison between the United Kingdom and the Netherlands on one hand and France and Belgium on the other demonstrates.
The strength of our comparative analysis is that it follows a standard approach for every country and uses the same basic model with the underlying distribution of the disease as a function of age. It allows for a better comparison across countries. Another advantage is that we are able to estimate an average investment per child per year per country. It is known that for every child born in one of the 4 countries studied, the health authorities will invest on average between €7.67 and €12.26 in direct medical costs against rotavirus disease per year and between €10.2 and €15.5 in indirect costs per year. The total costs (direct and indirect) are very similar for all 4 countries at €23.11 ± 0.7 (±3%).
The limitations of our analysis should also be considered. The cost per unit for each service offered and the total number of rotavirus disease events has been difficult to quantify with sufficient accuracy. The use of modeling implies that assumptions and uncertainties are introduced into the evaluation. Thus, our modeling approach may be open to criticism regarding the way some data are analyzed and interpreted, particularly with respect to the missing data. Serious disease events may lead to hospitalization whereas mild and moderate events are more often treated at home or in outpatient settings. In that respect, the model introduces limits on the units it generates: the number of emergency visits and hospitalizations should always be lower than the number of serious rotavirus-related AGE events. The sum of subjects staying at home, not seeking medical advice, together with those seeking advice should be equal to the total number of children with rotavirus diarrhea. Using sensitivity analysis, we tested these uncertainties to observe their combined importance as shown in Figure 3.
Our approach is appropriate or even conservative when the results presented here are compared with recent investigations in Belgium and France.42,43 For instance, the direct medical and grand total costs for rotavirus disease estimated by Bilcke et al42 for Belgium amount to €7.4 million and €19.6 million, respectively. Our results of €7.7 million for direct medical and €14.5 million (unadjusted cost figures in Tables 6 and 8, respectively) for the grand total were in line for the direct medical, but underestimated for the indirect cost as expected. For France, Huet et al43 reported total direct costs of €63 million to the National Healthcare Payer rising to €177 million from the societal perspective compared with €48.5 million and €87.1 million, respectively, here. This can in part be explained by the inclusion of out-of-pocket costs by Huet et al. Another important difference is that Huet et al reported a much higher percentage of children with rotavirus AGE seeking medical care compared with this publication. Lorgelly et al44 reported the total cost per child from a societal perspective in the United Kingdom as £86.33. During the at risk period of 5 years this would equate to approximately €26 per year, compared with the unadjusted total cost per child per year of €24 presented in this analysis.
Finally, the Netherlands will in general report a lower indirect cost than presented here as they are using the friction cost method45 to estimate this type of societal cost, which could be half the cost calculated when the human capital method is chosen.31 Our estimate for indirect costs should be substantiated with additional data to be collected through information supplied by the parents. We expect to have details on this type of data soon. In the absence of accurate data, the cost estimate for parental and sick child-related leave yielded only a range. In this analysis, it was further assumed that only mothers took sick child-related leave.
In healthcare systems such as those of the Netherlands and the United Kingdom, much of the financial responsibility of caring for children with rotavirus disease falls on the parents. Indirect costs are often not included in health economic analyses but constitute a considerable burden for families and society. Thus, in countries where first-line management is most promoted, a vaccination policy to prevent rotavirus disease is likely to benefit individuals and employers more than the healthcare sector. Employers may see the benefit in offering a free-of-charge vaccination program to young children of their employees (Table 9). Further data are required before this approach is more thoroughly considered.
Rotavirus infection is a major cause of AGE in children under the age of 5 years.2 Children with severe diarrheal disease are often hospitalized, contributing a significant cost to the healthcare expenditures in a country. In addition, rotavirus disease causes considerable burden for parents and families who must take time off work or other activities to care for their sick children. Studies in the United Kingdom and the United States have shown that rotavirus vaccination is a cost-effective intervention and can improve the QoL of children and their parents affected by rotavirus AGE.44,46
The authors thank Rachel Emerson who provided medical writing services.
1. Parashar UD, Bresee JS, Gentsch JR, Glass RI. Rotavirus
. Emerg Infect Dis
2. Parashar UD, Hummelman EG, Bresee JS, Miller MA, Glass RI. Global illness and deaths caused by rotavirus
disease in children. Emerg Infect Dis
3. Parashar UD, Gibson CJ, Bresee JS, Glass RI. Rotavirus
and severe childhood diarrhea. Emerg Infect Dis
4. Koopmans M, Brown D. Seasonality and diversity of Group A rotaviruses in Europe
. Acta Paediatr Suppl
5. Giaquinto C, Van Damme P, Huet F, Gothefors L, Van der Wielen M. Costs of community-acquired pediatric rotavirus
gastroenteritis in 7 European countries: the REVEAL Study. J Infect Dis.
6. Van Damme P, Van der Wielen M, Ansaldi F, et al. Rotavirus
vaccines: considerations for successful implementation in Europe
. Lancet Infect Dis
7. Djuretic T, Ramsay M, Gay N, Wall P, Ryan M, Fleming D. An estimate of the proportion of diarrhoeal disease episodes seen by general practitioners attributable to rotavirus
in children under 5 y of age in England and Wales. Acta Paediatr Suppl
8. Ryan MJ, Ramsay M, Brown D, Gay NJ, Farrington CP, Wall PG. Hospital admissions attributable to rotavirus
infection in England and Wales. J Infect Dis
. 1996;174(suppl 1):S12–S18.
9. Melliez H, Boelle PY, Baron S, Mouton Y, Yazdanpanah Y. [Morbidity and cost of rotavirus
infections in France]. Med Mal Infect
10. Brandt CD, Kim HW, Rodriguez WJ, et al. Pediatric viral gastroenteritis during eight years of study. J Clin Microbiol
11. Fourquet F, Desenclos JC, Maurage C, Baron S. [Acute gastro-enteritis in children in France: estimates of disease burden
through national hospital discharge data]. Arch Pediatr
12. Law A, Kelton W. Simulation Modeling and Analysis
. 3rd ed. Boston, MA: McGraw Hill Higher Education; 2000.
13. Giaquinto C, Van Damme P, Huet F, et al. Clinical consequences of rotavirus
acute gastroenteritis in Europe
, 2004–2005: the REVEAL study. J Infect Dis
. 2007;195(suppl 1):S26–S35.
14. Clemens J, Rao M, Ahmed F, et al. Breast-feeding and the risk of life-threatening rotavirus
diarrhea: prevention or postponement? Pediatrics
15. Newburg DS, Peterson JA, Ruiz-Palacios GM, et al. Role of human-milk lactadherin in protection against symptomatic rotavirus
16. Ruuska T, Vesikari T. Rotavirus
disease in Finnish children: use of numerical scores for clinical severity of diarrhoeal episodes. Scand J Infect Dis
17. Gleizes O, Desselberger U, Tatochenko V, et al. Nosocomial rotavirus
infection in European countries: a review of the epidemiology, severity and economic burden of hospital-acquired rotavirus
disease. Pediatr Infect Dis J
. 2006;25(1 suppl):S12–S21.
18. OECD in Figures 2006–2007. 2006. Available at: www.oecd.org/infigures
. Accessed July 6, 2007.
19. Kemmeren JM, Mangen M-JJ, van Duynhoven YTHP, Havelaar AH. Rotavirus, in Priority Setting of Food Borne Pathogens. Disease Burden and Costs of Selected Enteric Pathogens
. RIVM report 330080001/2006. Bilthoven: Rijksinstituut voor Volksgezondheid en Milieu (National Institute for Public Health and the Environment); 2006.
20. Official medical tariff list (UCANSS). [Ucanss 2006]. June 22, 2007. Available at: http://www.ucanss.fr/
. Accessed July 1, 2007.
21. Fruytier AC, Van Schoor J, Standaert B. Vaccination with RIX4414 is cost-effective in a Belgian setting (Poster PIH3). Presented at: ISPOR 9th Annual Congress; October 29, 2006; Copenhagen, DK.
22. Harris JP, Jit M, Cooper D, Edmunds WJ. Evaluating rotavirus
vaccination in England and Wales. I. Estimating the burden of disease. Vaccine
23. Martinot A, Dumonceaux A, Grandbastien B, Hue V, Leclerc F. Reseau interhospitalier d'evaluation des pratigues medicales dans les affections courantes de l'enfant. [Evaluation of the ambulatory treatment of acute diarrhea in infants]. Arch Pediatr
24. de Wit MA, Koopmans MP, van der Blij JF, van Duynhoven YT. Hospital admissions for rotavirus
infection in the Netherlands. Clin Infect Dis.
25. National Costs Study 2005. [Public hospitals]. 2005. Available at: www.atih.sante.fr
. Accessed October 16, 2007.
26. Contactisolatie kinderen (Contact isolation children). [Dutch Workingparty on Infection Prevention]. 2004. Available at: http://www.wip.nl/
. Accessed July 6, 2007.
27. Oostenbrink J, Koopmanschap M, Rutten F. Handleiding voor kostenonderzoek—methoden en standaard kostprijzen voor economische evaluaties in de gezondheidszorg
[Guide to Cost Research—Methods and Standard Costs Prices for Economic Evaluations in Healthcare
]. Amstelveen: College voor zorgverzekeringen (Health Care Insurance board); 2004.
28. Oostenbrink R, Moons KG, Derksen-Lubsen AG, Grobbee DE, Moll HA. A diagnostic decision rule for management of children with meningeal signs. Eur J Epidemiol
29. Sermet-Gaudelus I, de La RF, Salomon JL, et al. [Rotavirus
nosocomial infection in pediatric units. A multicentric observation study]. Pathol Biol (Paris)
30. Harrington M, Butler K, Cafferkey M. Rotavirus
infection in hospitalised children: incidence and impact on healthcare resources. Ir J Med Sci
31. Le T, Gibson J, Oxley L. Cost- and income-based measures of human capital. J Econ Surv
32. Directionate General of Human Rights. Parental Leave in Council of Europe
Member States. Strasbourg: Council of Europe
33. European Foundation for the Improvement of Living and Working Conditions. Parental leave in European companies: establishment survey on working time 2004–2005. Dublin, Ireland: European Foundation for the Improvement of Living and Working Conditions; 2007:1–57.
34. Moss P, O'Brien M. International review of leave policies and related research 2006. UK: Department of Trade and Industry; 2006. Employment relations research series no. 57.
35. Guide sécurité sociale de l'employeur. [L'Union des Caisses Nationales de Sécurité Sociale]. 2006. Available at: http://www.ucanss.fr/services/textes_documents/gsse/accueil_gsse.htm
36. Interruption de carrière dans le cadre du congé parental. [NEO - National employment office]. 2007. Available at: http://www.onem.fgov.be/D_opdracht_LBO/Regl/Werknemers/ouderschap/Ov_Cp/ContentFR.htm#TopOfPage
37. Time off for dependants: a guide for employers and employees (UK). [Department of Trade and Industry (UK)]. 2000. Available at: http://www.berr.gov.uk/files/file11419.pdf
38. Bevolking, geboorte. (Population, birth). [Centraal Bureau voor de Statistiek, Voorburg/Heerlen]. 2007. Available at: www.statline.cbs.nl
. Accessed July 9, 2007.
39. Mid-2005 population estimates. [Office of National Statistics]. 2007. Available at: http://www.statistics.gov.uk/statbase/Expodata/Spreadsheets/D9387.xls
40. Al MJ, Goossens L, Hövels AM, Michielsen CP, Standaert B. Cost-effectiveness of mass vaccination with a rotavirus
vaccine in the Netherlands (Poster PIN19). Presented at: ISPOR 9th Annual Congress; October 30, 2006; Copenhagen, DK.
41. Martin A, Standaert B. Cost effectiveness of infant vaccination with Rotarix in the UK. Presented at: 25th Annual Meeting of the European Society for Paediatric Infectious Diseases, ESPID; 2007; Porto, Portugal.
42. Bilcke J, Beutels P, De Smet F, Hanquet G, Van Ranst M, Van Damme P. Vaccination des nourrissons contre le rotavirus en Belgique. Analyse coût-efficacité
. KCE reports 54B. Bruxelles, Belgium: Centre fédéral d'expertise des soins de santé (KCE); 2007.
43. Huet F, Largeron N, Trichard M, Miadi-Fargier H, Jasso-Mosqueda G. Burden of paediatric rotavirus
gastroenteritis and potential benefits of a universal rotavirus
vaccination programme with RotaTeq in France. Vaccine
44. Lorgelly PK, Joshi D, Iturriza GM, Gray J, Mugford M. Exploring the cost effectiveness of an immunization programme for rotavirus
gastroenteritis in the United Kingdom. Epidemiol Infect.
45. Brouwer WB, Koopmanschap MA. The friction-cost method: replacement for nothing and leisure for free? Pharmacoeconomics
46. Widdowson MA, Meltzer MI, Zhang X, Bresee JS, Parashar UD, Glass RI. Cost-effectiveness and potential impact of rotavirus
vaccination in the United States. Pediatrics