In the past 2 decades, there has been no appreciable change in the incidence of diarrheal disease worldwide, including disease caused by rotavirus,1 but the number of deaths attributable to diarrhea has approximately halved as a result of improved nutrition and access to health care, particularly oral dehydration therapy.1,2 Deaths caused by diarrheal illness, including rotavirus, occur infrequently in industrialized regions such as Europe. Yet despite the low number of deaths annually, the burden of disease due to rotavirus remains considerable, although not precisely known. A model developed by the Centers for Disease Control and Prevention (CDC) to evaluate the global burden of rotavirus disease estimated that >220,000 hospitalizations and 1.7 million outpatient visits were attributable to rotavirus each year in industrialized countries, including Europe, the United States, Japan and Australia.2 A further 7 million cases of rotavirus infection were managed entirely at home.
There are few prospective studies describing the burden of disease caused by rotavirus in European Union (EU) countries.3–10 In some prospective studies, ∼4/100 children 48 months of age or younger presented each year to a physician with rotavirus gastroenteritis in Germany,3,4 compared with 0.84/100 in Austria, 1.38–1.6/100 in Switzerland,4,9 0.49/100 in the Netherlands11 and 2.8/100 children in Finland.12 Prospective studies suggest possible regional differences, with cumulative hospitalization incidence rates ranging between 1 in 17 and 1 in 95 (median 1 in 67), but there are insufficient data to confirm these findings (J. Wolleswinkel-van den Bosch and C. Giaquinto, unpublished observations). In most countries, detection of rotavirus is not notifiable, and disease estimates are frequently based on laboratory data or hospital discharge data. The incidence of both community-acquired and nosocomial rotavirus infections might be underestimated when based on hospital discharge data.13,14 Furthermore, because detection of rotavirus in stool samples does not influence treatment, there is little incentive for physicians to collect stool samples from patients presenting with gastroenteritis and many cases can go undiagnosed. Both factors imply that the incidence of community-acquired and nosocomial rotavirus infection may be substantially underestimated.
A live attenuated G1P human rotavirus vaccine (Rotarix™; GlaxoSmithKline Biologicals, Rixensart, Belgium) has been licensed in Mexico and several Latin-American, Middle East, Asian and African countries and has been submitted for licensure within the EU.15 A live human-bovine reassortant G1, G2, G3, G4 and P1A vaccine (RotaTeq®; Merck & Co., Inc., Whitehouse Station, NJ) has been recently submitted for licensure in the United States.16 Current, accurate data describing the epidemiology and burden of rotavirus infections are needed to guide future recommendations for vaccine use.
To provide an overall estimate of rotavirus disease within the expanded EU, we applied the model based on the global approach developed by the CDC to estimate the burden of rotavirus disease in each of the 25 EU member states. We added up these country-specific estimates to obtain a global EU estimate and made a distinction between rotavirus episodes at home, clinic visits, hospitalizations and deaths caused by rotavirus.
Nonfatal Rotavirus Infections.
The initial CDC model estimated the rotavirus hospitalization rate from published studies in industrialized countries. Studies were included if they conformed to quality criteria, including a study duration of at least 1 year, inclusion of children younger than 5 years of age, and use of reliable rotavirus detection assays. Ten studies performed in Europe, Australia and the United States were included in the CDC model to give an overall rotavirus hospitalization rate of 445/100,000 for children younger than 5 years of age living in industrialized countries. We excluded data from the 2 Australian and 1 United States study because they might not apply directly to the European context, and recalculated the yearly number of rotavirus hospitalizations among children younger than 5 years of age in EU countries. The recalculated rate was 370/100,000. We then assumed, following CDC model's assumptions based on published studies from Chile, Finland and Canada,12,17,18 that for each hospitalization, ∼8 children visit a physician. To estimate the number of episodes treated at home, the model assumes that for every rotavirus infection leading to a physician visit, 4 episodes occur at home.
Deaths Caused by Rotavirus.
EU countries were grouped according to per capita gross national income (GNI) according to the 2003 World Bank classification of income.19 Economies were divided into upper middle income ($3036–9385) and high income ($9386 or more). For each individual country, the total number of deaths reported in children younger than 5 years of age was recorded from United Nations (UN) statistics (most current year available20). For countries where vital statistics data were not available for children younger than 5 years of age (Table 1, Cyprus), we estimated the population from UNICEF year 2000 data, and the number of deaths was estimated proportionally to the number of deaths in a country representative of the same GNI group (Germany high GNI group). The proportion of deaths attributable to diarrhea was estimated by Parashar2 based on published studies and vital statistics and was calculated to be 9 and 1% in each income group, respectively. To estimate the number of deaths caused by rotavirus, the proportion of hospitalizations for diarrhea caused by rotavirus (assumed to be 31 and 34% for the same respective GNI groups) was multiplied by the diarrhea-caused mortality rate of children younger than 5 years of age for each country. This provided an estimation of the number of deaths caused by rotavirus in each country.
Nonfatal Rotavirus Infections.
According to recent UN and UNICEF data, the total population younger than 5 years of age in the 25 EU member states is ∼23,598,000 (data from years 2000–2003). Using an estimate of 370 (range, 290–565), children hospitalized for rotavirus per 100,000 children and multiplying by the total population younger than 5 years of age in each of the 25 EU countries, it is estimated that 87,313 (range, 68,400–133,300) hospitalizations caused by rotavirus occur each year in children younger than 5 years of age in the 25 EU member states (Table 1 and Fig. 1).
It can be assumed that for each child hospitalized, ∼8 children (range, 5–10) require a visit to a health care facility or physician's office, and that for each child requiring medical care, ∼4 children (range, 3–5) require only home care. On this basis, we estimated 698,501 cases (range, 436,500–873,100) of rotavirus disease seen in clinics and 2,794,003 (range, 2095,500–3,493,000) rotavirus cases treated at home each year in the EU.
Deaths Caused by Rotavirus.
The income group for each country is listed in Table 1. We estimate that 231 deaths caused by rotavirus disease occur annually in children younger than 5 years of age within the EU. The number of deaths estimated for countries from a high income group is quite small, ranging from 1/39,900 in Malta to 1/54,000 in Germany. For the upper middle income group, the estimated number of deaths is still considerable, ranging from 1/2900 in Latvia to 1/6800 in the Czech Republic.
The availability of effective vaccines against rotavirus disease has stimulated efforts to characterize the burden of disease caused by this pathogen. By using a previously developed model based on published data and adapting this model to current country-specific population and mortality data, we have estimated the global health burden resulting from rotavirus disease in the EU. We estimate that ∼3.6 million diarrheal episodes occur each year because of rotavirus infection in children younger than 5 years of age. This translates to 1 symptomatic infection in every 7 children each year. This is close to an estimate from a prospective Finnish study in which symptomatic infection was observed in ∼1 in every 6 children younger than 18 months of age.8 Most rotavirus infections are mild enough to be managed at home, but ∼20% of children will see a medical practitioner during the illness, and an estimated 1 in 54 cases will result in hospitalization.
The difference in epidemiology between countries with high-middle and high GNI according to World Bank criteria is illustrated in Figure 1. The model assumes a similar hospitalization rate across GNI groups, whereas actual differences in under 5-year mortality are reflected in the predicted death rates caused by rotavirus disease. Overall mortality of children younger than 5 years of age is reported, for example, as 1.1 deaths/1000 children in Germany (high GNI), 1.78/1000 in Poland (high-middle GNI) and 2.5/1000 in Latvia (high-middle GNI). The large difference in the estimated number of rotavirus deaths is thus largely the result of a difference in overall mortality rates in children and the assumption that a 9-fold higher proportion of deaths is caused by diarrhea in the low-middle income group of countries compared with the high income group. We are aware that these estimates are highly dependent on assumptions that are difficult to check and that might be less robust than the other estimates.
The rates we have estimated for the burden of rotavirus disease in Europe are not dissimilar to published rates of morbidity and death caused by rotavirus in the United States. It is estimated that >2.7 million cases of rotavirus occur in children younger than 5 years of age in the United States each year, resulting in 570,000 outpatient visits, 50,000 hospitalizations and 20–40 deaths, with a cumulative risk to age 5 years of 1:7, 1:78 and 1:200,000, respectively (birth cohort of 3.9 million).21–23 Although the cumulative risk of death appears to be lower in the United States than in other high GNI countries within Europe, one author noted that because only deaths in hospital were considered, the number of estimated deaths caused by to rotavirus is likely to be an underestimate because it does not consider children who might have died outside of, or before reaching, the hospital.24
Few studies have attempted to measure incidence rates of rotavirus diarrhea that do not result in medical attention. The model uses the available estimates from Finland, Canada and Chile12,17,18 to derive the number of cases of rotavirus illness seen in clinics or managed at home. The model does not take into account the possibility of multiple visits to medical practitioners, nor does it differentiate between visits to the physician or emergency departments. Health care utilization and medical management of rotavirus infection have been shown to differ in adjacent countries within Europe.4 In Germany, 12% of children with rotavirus diarrhea visited a physician on at least 4 occasions during their illness, compared with 2% in Austria and Switzerland. In the same study, differences (not statistically significant) in hospitalization rates across 3 countries (7.3, 4 and 11.5% in Germany, Austria and Switzerland, respectively) were not explained by differences in disease severity between countries.14 Health care behaviors and management practices, such as parents’ tendency to seek medical advice, as well as the criteria leading to hospital admission, all impact on health care utilization, underscoring the importance of country-specific epidemiologic data to understand the health care burden of rotavirus locally. We have not integrated this variability into our model.
Nosocomial infections are not considered in this model. In industrialized countries, nosocomially acquired rotavirus affected approximately the same number of children as community-acquired illness.14 Rotavirus accounted for up to 40% or more of nosocomially acquired diarrhea in Poland, Austria and France5,10,25 with attack rates between 11 and 15%.28–31 Two studies in France and Italy demonstrated that a further 15% of children become symptomatic after discharge.28,29 Costs associated with nosocomial infections are considerable and can account for the majority of hospital-associated costs for rotavirus disease in industrialized countries.5 The economic burden of nosocomial rotavirus infection in Europe is presented in the article by Gleizes et al32 in this supplement.
Rotavirus is the single most common cause of gastroenteritis in children younger than 5 years of age and is 3 times more likely to result in hospitalization than gastroenteritis resulting from other causes.3 In industrialized as well as developing countries, there is no evidence to support a reduction in the frequency of diarrhea caused by rotavirus in the past 20 years.1,12 Although death rates have fallen, improvements in sanitation and hygiene have not changed the global incidence of disease, pointing to vaccination as the most effective way to achieve disease control. New rotavirus vaccines are under development, and 2 are close to introduction in the EU. Both the human vaccine developed by GlaxoSmithKline Biologicals and the bovine vaccine developed by Merck have shown high efficacy in clinical trials33–35 and have a good safety profile.36,37
Although our study highlights the extent to which rotavirus impacts on the health of children and on health care costs in Europe, it is unable to provide the country-specific information likely to be required for formulation of vaccination policy. This study draws attention to the need for prospective studies to assess country-specific epidemiology of rotavirus infection. When new rotavirus vaccines become available, rotavirus will be the single most frequent vaccine-preventable illness affecting infants and young children in the EU.
We thank Dr Umesh Parashar for his input to the model and Drs Jo Wolter and Carlo Giaquinto for their contributions and comments on the manuscript.
1.Kosek M, Bern C, Guerrant RL. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull World Health Organ
2.Parashar UD, Hummelman EG, Bresee JS, et al. Global illness and deaths caused by rotavirus disease in children. Emerg Infect Dis
3.Ehlken B, Laubereau B, Karmaus W, et al. Prospective population-based study on rotavirus disease in Germany. Acta Paediatr
4.Frühwirth M, Heiniger U, Ehlken B, et al. International variation in disease burden of rotavirus gastroenteritis in children with community- and nosocomially acquired infection. Pediatr Infect Dis J
5.Frühwirth M, Berger K, Ehlken B, et al. Economic impact of community- and nosocomially acquired rotavirus gastroenteritis in Austria. Pediatr Infect Dis J
6.Takala AK, Koskenniemi E, Joensuu J, et al. Economic evaluation of rotavirus vaccinations in Finland: randomized, double-blind, placebo-controlled trial of tetravalent rhesus rotavirus vaccine. Clin Infect Dis
7.Ruuska T, Vesikari T. A prospective study of acute diarrhoea in Finnish children from birth to 2.5 years of age: clinical severity, etiology, and risk factors. Acta Paediatr Scand
8.Joensuu J, Koskenniemi E, Vesikari T. Randomised placebo-controlled trial of rhesus-human reassortant rotavirus vaccine for prevention of severe rotavirus gastroenteritis. Lancet
9.Laubereau B, Gateau S, Ehlken B, et al. Rotavirus-Gastroenteritis im Säuglings- und Kleinkindesalter. Schweiz Med Wochenschr
10.Thuret A, Patural H, Berthelot P, et al. Prospective follow-up of hospital-acquired diarrhoea in 28 paediatric wards of the south-east part of France during a winter season. Pathol Biol
11.Koopmans M, Van Asperen I. Epidemiology of rotavirus infections in the Netherlands. Acta Pediatr Suppl
12.Vesikari T, Rautanen T, Von Bonsdorff CH. Rotavirus gastroenteritis in Finland: burden of disease and epidemiological features. Acta Paediatr Suppl
13.Hsu VP, Staat MA, Roberts N, et al. Use of active surveillance to validate international classification of diseases code estimates of rotavirus hospitalizations in children. Pediatrics
14.Berner R, Schumacher RF, Hameister S, et al. Occurrence and impact of community-acquired and nosocomial rotavirus infections: a hospital-based study over 10 y. Acta Paediatr Suppl
15.Verstraeten T, Wolleswinkel-van den Bosch JH. Burden of Rotavirus Disease in the New European Union. Annual Meeting of the European Society for Paediatric Infectious Diseases (ESPID) May 8–20, 2005, Spain.
17.Rodriguez WJ, Kim HW, Brandt CD, et al. Longitudinal study of rotavirus infection and gastroenteritis in families served by a pediatric medical practice: clinical and epidemiologic observations. Pediatr Infect Dis J
18.Ford-Jones EL, Wang E, Petric M, et al. Rotavirus-associated diarrhea in outpatient settings and child care centers: The Greater Toronto Area/Peel Region PRESI Study Group. Pediatric Rotavirus Epidemiology Study for Immunization. Arch Pediatr Adolesc Med
21.Tucker AW, Haddix AC, Bresee JS, et al. Cost-effectiveness analysis of a rotavirus immunization program for the United States. JAMA
22.Rotavirus Vaccine for the Prevention of Rotavirus Gastroenteritis Among Children: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR
23.Kilgore PE, Holman RC, Clarke MJ, et al. Trends of diarrheal disease-associated mortality in US children, 1968 through 1991. JAMA
24.Jin S, Kilgore PE, Holman RC, et al. Trends in hospitalizations for diarrhea in United States children from 1979 through 1992: estimates of the morbidity associated with rotavirus. Pediatr Infect Dis J
25.Mrukowicz JZ, Krobicka B, Duplaga M, et al. Epidemiology and impact of rotavirus diarrhoea in Poland. Acta Paediatr Suppl
26.Deleted in proof.
27.Deleted in proof.
28.Piednoir E, Bessaci K, Bureau-Chalot F, et al. Economic impact of healthcare-associated rotavirus infection in a paediatric hospital. J Hosp Infect
29.Gianino P, Mastretta E, Longo P, et al. Incidence of nosocomial rotavirus infections, symptomatic and asymptomatic, in breast-fed and non-breast-fed infants. J Hosp Infect
30.Bennet R, Hedlund K-O, Ehrnst A, et al. Nosocomial gastroenteritis in two infant wards over 26 months. Acta Paediatr
31.Johansen K, Bennet R, Bondesson K, et al. Incidence and estimates of the disease burden of rotavirus in Sweden. Acta Paediatr Suppl
32.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
33.Vesikari T, Karvonen A, Puustinen L, et al. Efficacy of RIX4414 live attenuated human rotavirus vaccine in Finnish infants. Pediatr Infect Dis J
34.De Vos B, Vesikari T, Linhares AC, et al. A rotavirus vaccine for prophylaxis of infants against rotavirus gastroenteritis. Pediatr Infect Dis J
35.Vesikari T, Matson D, Dennehy P, et al. Protection against rotavirus gastroenteritis of multiple serotypes by a pentavalent (human bovine) reassortant rotavirus vaccine (PRV). In: 23rd Annual Meeting of the European Society for Paediatric Infectious Diseases (ESPID)
.Valencia, Spain: European Society for Paediatric Infectious Diseases; 2005.
36.Vesikari T, Matson D, Van Damme P, et al. Incidence of intussusception with the pentavalent (human-bovine) reassortant rotavirus vaccine (PRV) is similar to placebo. In: 23rd Annual Meeting of the European Society for Paediatric Infectious Diseases (ESPID)
. Valencia, Spain: European Society for Paediatric Infectious Diseases; 2005.
37.Vesikari T, O'Ryan M, Abate H, et al. Overcoming the safety hurdle: the rotavirus vaccine RIX4414 is not associated with intussusception. In: 23rd Annual Meeting of the European Society for Paediatric Infectious Diseases (ESPID)
. Valencia, Spain: European Society for Paediatric Infectious Diseases; 2005.