Eur J Pediatr (2005) 164: 607–615 DOI 10.1007/s00431-005-1705-0

O R I GI N A L P A P E R

Birgit Ehlken Æ Gabriele Ihorst Æ Barbara Lippert Angela Rohwedder Æ Gudula Petersen Martin Schumacher Johannes Forster Æ for the PRIDE Study Group

Economic impact of community-acquired and nosocomial lower respiratory tract infections in young children in Germany

Received: 3 January 2005 / Revised: 2 May 2005 / Accepted: 3 May 2005 / Published online: 18 June 2005 Ó Springer-Verlag 2005

Abstract Data on the economic burden of lower respiratory tract infections (LRTI) in young children are lacking in Germany. The objective of the cost-of-illness study was to estimate the economic impact of community-acquired LRTI and nosocomial LRTI as well as of infections due to respiratory syncytial virus (RSV), parainfluenza viruses (PIV) and influenza viruses (IV). The economic analysis is part of the PRI.DE study, a prospective, multi-centre, population-based epidemiological study on the impact of LRTI in children aged 0 to 36 months in Germany. The analysis includes children with community-acquired infections (1329 cases treated as outpatients, 2039 cases treated as inpatients) and nosocomial infections (90 cases). Medical services consumed were generated by chart abstraction and parental expenses data by telephone interviews within four weeks after physician visit or hospitalisation. Costs were evaluated from following perspectives: third party payer, parent and society. Total costs for outpatient treatment are €123 per LRTI case. Stratified by virus type, total costs per case are €163 (RSV), €100 (PIV) and €223 (IV). B. Ehlken (&) Æ B. Lippert MERG, Medical Economics Research Group, Munich, Germany E-mail: [email protected] G. Ihorst Æ M. Schumacher IMBI, Institute of Medical Biometry and Medical Informatics, University Hospital Freiburg, Freiburg, Germany A. Rohwedder Institute for Medical Microbiology and Virology, Ruhr-University, Bochum, Germany G. Petersen Wyeth, Muenster, Germany J. Forster St. Joseph’s Hospital Freiburg, Freiburg, Germany G. Ihorst Center for Clinical Trials, University Hospital Freiburg, Freiburg, Germany

Total costs per hospitalised LRTI case amount to €2579. Stratified by virus type, total costs per case are €2772 (RSV), €2374 (PIV) and €2597 (IV). Total costs per nosocomial case are €2814. Economic burden due to LRTI is €213 million annually. It is concluded that treatment of LRTI in children up to age three causes a considerable economic burden in Germany. Presented results are the first data describing the economic burden of LRTI in young children assessed by means of the incidence data for Germany. This cost-of-illness study provides basic data for further decision-making, focusing on the economic assessment of preventive strategies for RSV, PIV and IV infections. Keywords Cost-of-illness Æ Respiratory tract infection Æ Respiratory syncytial viruses Æ Parainfluenza viruses Æ Influenza viruses List of Abbreviations CI: Confidence interval Æ CRF: Case report form Æ IV: Influenza virus Æ LRTI: Lower respiratory tract infections Æ OTC: Over the counter Æ PIV: Parainfluenza virus Æ RSV: Respiratory syncytial virus Æ TPP: Third party payers

Introduction Respiratory syncytial virus (RSV) is the major respiratory pathogen of young children and the foremost cause of severe lower respiratory disease in infants [1, 7, 10]. Infection with RSV is seen throughout the world in annual epidemics that occur in late fall, winter or spring; re-infection is common. RSV accounts for 20–25% of hospital admissions of young infants and children for pneumonia and for up to 75% of cases of bronchiolitis in this age group. Rates of illness are highest among infants between one and six months of age and more than 90% of all children are infected by two years of age. Premature infants are more likely to be infected with

608

RSV than infants born at term and have a higher ongoing morbidity [3, 8, 9]. The efficacy of RSV prophylaxis by application of a monoclonal antibody or immunoglobulin in high-risk infants has been shown in recent studies but not for a broader selected population [13, 23, 25, 26]. The control of RSV infection through preventive and prophylactic strategies, such as the development of vaccines, remains a major goal. Parainfluenza viruses (PIV) are known to account for a large percentage of paediatric respiratory infections including laryngotracheobronchitis (croup), bronchiolitis and pneumonia. Parainfluenza viruses are the major cause of croup. Children between the ages of 3 and 36 months are commonly affected by parainfluenza related laryngotracheobronchitis. Reinfections are common. Time period between PIV 3 infections was reported to be on average 1.4 years (ranging from six months to three years) in children younger than five years [16]. Symptomatic treatment of a parainfluenza infection is available but as yet, no specific antiviral agents are available. Influenza viruses (IV) attack the respiratory tract. Most people recover in one or two weeks, but some, especially the elderly and very young children, develop complications such as pneumonia, bronchitis and ear infections or even life-threatening complications, including kidney and heart failure. A US study including healthy children reported comparable hospitalisation rates for illness attributable to influenza in children younger than six months and adults at high risk for influenza [15]. German epidemiological data indicate that the burden of influenza hospitalisations was about one-third that of RSV [28]. Besides treatment of symptoms, an influenza vaccine is available, which, however, due to the high mutation rate of the virus, must be given annually. Data describing the economic burden of lower respiratory tract infections (LRTI) in infants and children due to RSV and PIV in Germany are not yet published. Socio-economic data regarding influenza exist, but focus mainly on the whole population or the elderly [20, 24]. Economic data regarding children are lacking in Germany. The aim of this evaluation was to describe the economic burden of LRTI as well as for infections due to RSV, PIV and IV in children up to age three years in outpatient and inpatient care in Germany. First, average costs are shown per LRTI case and with infections caused by RSV, PIV and IV. Second, the economic impact of LRTI and due to RSV, PIV and IV is assessed by means of the incidence data for Germany.

study conducted from 1999 to 2001 on the impact of PIV and RSV infections in children aged 0 to 36 months in Germany. The study was designed aiming to represent the German population of children of less than three years by a) multi-centre sampling (one city each in the north, east, south and west of the country) and b) recruiting children at each site in paediatric practices and in referral children’s hospitals. This procedure assured near-complete ascertainment, because in the respective age group, 95% of children in Germany have a paediatrician and in case of a hospitalisation, more than 95% are referred to children’s hospitals (unpublished data from German Association of Paediatricians). The study protocol was approved by the Ethics Committee of the University Hospital Freiburg, and by the Ethics Committee of each participating hospital. All direct medical and non-medical costs were evaluated from the perspective of third party payers (TPP), as well as parent and family perspectives. In addition, indirect costs were determined and finally, costs were evaluated from the societal perspective. Recruitment Subjects were recruited by 11 office-based paediatricians (2–3 practices in each city) and 5 hospitals in 4 different regions in Germany (Bochum, Dresden, Freiburg, Hamburg). Office-based physicians were asked to enrol consecutively up to 250 patients fulfilling the inclusion criteria and hospital-based physicians up to 540 patients with community-acquired LRTI infection and all nosocomial LRTI infections (inpatients for more than 48 h when developing signs and symptoms of LRTI) within the given study period. Every child up to the age of 36 months presenting with a diagnosis meeting the case definitions was asked to participate in the study. If informed consent from the parents was not obtained, the patient was listed with the date of physician visit, sex, month of birth, diagnosis (case definition) and reason for refusal. In those cases where informed consent was given, a sample of nasopharyngeal aspirate (NPA) was taken and tested for virus load. Parents who gave their informed consent were also asked to participate in a telephone interview. In addition, data were obtained from the patients’ records. Data basis covers an observation period of two years (November 1, 1999 through October 31, 2001). Criteria for inclusion

Subjects and methods Research design The economic analysis is part of the PRI.DE study [6]. PRI.DE is a prospective, multi-centre, epidemiological

Infants and children up to the age of 36 months with lower respiratory tract infection (pneumonia, obstructive bronchitis, bronchitis, bronchiolitis, croup, and in infants less than 6 months, with apnoea) were included in the study. Written informed consent was obtained from the parents.

609

Data collection Physicians at the hospitals or trained medical staff at the paediatric practices collected utilization data. The following data were documented from patients’ records using a standardized case report form (CRF): medical services including consultations, physical examination and diagnostic procedures such as X-ray, prescribed medications and hospitalisation. Data regarding parental expenses and impact of LRTI on family life were collected by telephone interview with the parents within four weeks after the physician visit or hospitalisation. The following data were collected using a standardized patient questionnaire: demographic characteristics, ad-hoc expenses (e.g. OTC medications), travel expenses (public and private transport), co-payments (e.g. for inpatient care), auxiliary devices, non-reimbursed therapies, expenses for childcare and workdays lost, missed day care or foster family, missed appointments of the child, missed usual activities of the family, disturbances of mother’s and father’ sleep. To determine the type of virus (RSV A and B, PIV 1– 3, IV A and B), nasopharyngeal secretions were taken by a standardized procedure and tested using a PCR-test kit (Prodesse; USA) at a central testing laboratory. Costing The economic burden of LRTI is caused by direct medical costs (e.g. health services used for treatment) and direct non-medical costs (e.g. caregiver costs and travel expenses). Indirect costs from loss of work days by the parents or caregivers are also of interest. To evaluate costs from the perspective of third party payers, the average frequency a diagnostic or treatment service was used was calculated and multiplied by the charge. Charges for outpatient services were based on the points appropriated to the particular treatment published in the German tariff list, multiplied by the average value per point for medical services: €0.048 (2002) [5]. The prescribed medications were docu-

mented from patients’ records and multiplied by the reimbursable prices in the German pharmaceutical index [19]. These figures were then reduced by patients’ co-payments. Inpatient costs were calculated using an average of the daily rate reimbursed by German TPP for hotel costs and medical services (per diem rate: regular paediatric ward: €323.15 and intensive care unit: €886.41) [30]. The human capital approach was applied to calculate indirect costs arising from lost productivity [11]. Calculated from the data on gross income and number of employees in the Statistical Yearbook of 2001, the monetary value of productivity for employees was €93.74 per day [21]. Costs for LRTI-related loss of work were calculated exclusively for parents/caregivers employed during the observation period. Parents retrospectively estimated their own expenses for OTC medication, auxiliary devices and non-reimbursable therapies due to LRTI of their child. To determine the travel expenses of the parents, the calculation was based on the reimbursed costs per km by the statutory sickness funds (€0.22 per km). Epidemiological data Based on the annual incidences of community-acquired LRTI and hospitalisations due to LRTI as well as nosocomial LRTI (taken from the results of the epidemiological part of the PRI.DE study [6], Table 1) the annual economic burden of LRTI caused by RSV, PIV and IV in Germany was estimated using the median value of total cost for evaluation. The incidence of LRTI cases is based on the number of children years (zero to three years) under observation at the participating practices (k=8132) and on the number of children up to the age of three years living in Germany in 1999 (N=2,374,600), as published by Statistisches Bundesamt (2001) [12, 21]. Figures for the total number of hospitalisations in the German paediatric population were obtained by relating the proportion of LRTI cases for hospitalisations in the five study hospitals to the total number of hospitalisations in the German paediatric population up to the age of three years in 1999 (N=489,816) [22].

Table 1 Annual incidence of lower respiratory tract infection (LRTI) in children zero to three years old, stratified by virus type detected, PRI.DE paediatric practices and hospitals, 1999–2001 (mean, 95% conf. interval) Parameter

All LRTI

RSV

PIV

IV

Cases at paediatric practices (per 100 children and year) Absolute figure for Germany (cases per year) Hospitalisations (per 100,000 children and year) Absolute figure for Germany (hospitalisations per year) Nosocomial cases (per 100,000 hospital days)

28.7 (27.7–29.7)

7.7 (6.7–8.9)

3.7 (3.1–4.5)

1.1 (0.8–1.5)

682,128 (658,814–705,805) 183,761 (159,465–210,448) 88,994 (72,831–107,685) 25,977 (18,066–36,238) 2941 (2843–3042)

1117 (1003–1239)

261 (212–320)

123 (90–163)

69,847 (67,516 –72,228)

26,524 (23,812–29,432)

6205 (5043–7605)

2913 (2146–3873)

79 (66–93)

29 (17–46)

9.0 (3.5–19.2)

1.5 (0.1–7.1)

Data source: Forster et al. Eur J Pediatr 2004;163:709–716

610

Statistics The analysis was performed separately for the three study groups: practice cases, hospitalised cases and nosocomial cases. The SAS system (Version 8.2) was used for statistical analysis. Average cost per patient (case) was stratified by age group (0–12 months, 13–24 months, 25–36 months), type of virus (RSV, PIV, IV) and diagnosis (apnoea, bronchiolitis, pneumonia, obstructive bronchitis, bronchitis, croup).

Results Patient sample In total, 2389 eligible cases with LRTI, 2974 eligible cases hospitalised for LRTI and 141 nosocomial cases were observed at participating paediatric practices and hospitals. Data from a total of 1487 cases from paediatric practices (62.3%), 2068 hospitalised cases (70.7%) and 112 nosocomial cases (79.4%) with informed consent were collected; case children included in the study were similar (distribution of sex, age groups, diagnosis) to eligible children who declined to participate [6]. The main cause for non-participation was the high workload in practices and hospitals. A total of 1329 practice cases, 2039 hospitalised and 90 nococomial cases could be included in the economic analysis. Patients were excluded in the case information about resource utilization (e.g. length of stay in hospital was not available). Follow-up telephone interviews were carried out in 654 of the officebased cases, in 1118 of the hospitalised and 36 of the nosocomial cases. Demographic data for office-based, hospitalised, and nosocomial cases are summarized in Table 2. Resource utilization Table 3 depicts resource utilization, such as consultations of the office-based cases, length of stay of hospitalised cases, prolongation of stay in hospital due to LRTI in nosocomial cases, and days absent from work of parents or caregivers. The vast majority of children treated by office-based paediatricians (97%) received prescriptions for treatment, including inhalation, mucolytica, and antibiotics. Approximately 2% of the cases were admitted directly to the hospital by the physicians. In 4% of the cases, a Table 2 Demographic characteristics

chest X-ray was performed for diagnosis. Emergency transport was not necessary in any of the cases. Resource utilization from the parents and family perspectives (Table 4) was determined by expenses for supplemental childcare, transportation and additional expenses in all three cohorts. Costs for supplemental childcare played a considerable role for parents of children with nosocomial LRTI in comparison to office-based and hospital cases. The proportion of parents/caregivers who had lost workdays due to LRTI in their infant or child was approximately 14% in the office-based cohort, 8% in the cohort of hospitalised cases and 28% in the cohort of nosocomial cases.

Direct and indirect costs Overall, direct medical costs for TPP were €70 (SD 37) per office-based LRTI case (n=1329). The cost drivers included costs for physician visits and physical examinations (59%), followed by drug costs, which were responsible for 40% of these total direct medical costs. Direct medical costs per hospitalised LRTI case (n=1035) was €2,461 (SD 2088) for the TPP. Cost driver of direct medical costs for inpatient care was care on a regular paediatric ward and accounted for approximately 95% of direct medical costs. Care in the intensive care unit accounted for 5% of direct medical costs. Direct medical costs per nosocomial case (n=90) amounted to €3433 (SD 3815); treatment in the intensive care unit incurred 53% of total costs and on the regular paediatric ward, 47% of the costs. Direct non-medical costs for parents and caregivers amounted to €12 (SD 37) for supplemental childcare (50%) and travel expenses (50%) of office-based cases (n=609). Parents and caregivers of hospitalised cases incurred an average of €97 (SD 170) for additional expenses such as meals. A total of 61% of direct non-medical costs were for transport to and from the hospital (n=1049). For nosocomial cases, expenses amounted to €113 (SD 165) per case (n=27). Comparison of direct medical costs and indirect costs revealed that the latter were a major component of total costs only for office-based cases and not for hospital cases. €45 (SD 151) per office-based case (n=654), €136 (SD 283) per hospitalised case (n=1117) and €161 (SD 318) per nosocomial case (n=35) were due to lost work days of parents and caregivers.

Parameter

Office-based cases (n=1329)

Hospitalised cases (n=2039)

Nosocomial cases (n=90)

Sex (male) in % (n) Age groups 0–12 months in % (n) 13–24 months in % (n) 25–36 months in % (n) Mean age in years (SD)

56.3 (748)

60.7 (1237)

58.9 (53)

40.9 (543) 36.5 (485) 22.7 (301) 1.3 (0.8)

59.8 (1220) 28.1 (573) 12.1 (246) 1.0 (0.7)

91.1 (82) 0.8 (7) 1.1 (1) 0.4 (0.4)

611 Table 3 Medical resource utilization per case and days absent from work due to LRTI for office-based, hospitalised and nosocomial cases Parameter

Office-based cases mean (SD)

Hospitalised cases mean (SD)

Nosocomial cases mean (SD)

Number of consultations to paediatrician Number of emergency transports Length of hospital stay (days) Normal paediatric ward (days) Intensive care unit (days) Days absent from work (days)

1.7 (0.8)a 0.0 (0.0)a – – – 0.5 (1.6)b

– – 7.4 7.2 0.2 1.5

– – 7.2 5.2 2.0 1.7

c

(6.1) (5.2)c (1.4)c (3.0)d

(5.6)e (5.1)e (4.5)e (3.4)f

a

Patient charts (n=1329) Telephone interviews (n=654) Patient charts (n=2039) d Telephone interviews (n=1117) e Patient charts (n=90) f Telephone interviews (n=36) b c

Total costs Total costs from the societal perspective of office-based and hospitalised LRTI cases, stratified by age group, type of virus, and diagnosis are shown in Table 5. Total cost per case was analysed for all cases which had complete data regarding all cost components (direct medical and non-medical cost, indirect cost). Mean values for direct medical, direct non-medical and indirect cost in this sub-sample are nearly identical with the mean values based on all available cases for the respective cost components. In 568 office-based cases, total costs could be calculated. Total costs per case, considering direct medical and non medical costs as well as indirect costs per case, were €123 (SD 161). Close to 54% were direct medical costs, 11% were direct non-medical costs and 35% indirect costs. In the hospital cohort (n=1035), total costs from the societal perspective amounted to €2579 (SD 1874). Of the total cost, 91% was incurred by direct medical costs for hospitalisation. Expenses of the parents (direct nonmedical costs) were responsible for 4% of the average total cost and 5% of total costs were due to indirect costs. Total costs could be calculated for 22 nosocomial cases (€2814, SD 2944). Due to the small number, costs were not further stratified. Stratification of total costs by age group, type of virus and diagnosis

Annual economic burden due to lower respiratory tract infections

In the office-based setting total cost per case rose with increasing age. Whereas the direct medical and nonmedical costs showed no difference between the age groups, indirect cost rose with increasing age. Table 4 Percentage and number of patients with additional expenses and absence from work due to LRTI for office-based, hospitalised and nosocomial cases (based on telephone interviews)

In the hospital setting total costs per case decreased by increasing age. This was caused by the development of direct medical costs. Infants (0–12 months) were the most cost-intensive age group. Stratification of total costs by type of virus showed that practice cases with IV had the highest total costs followed by RSV and PIV. Of the total costs per IV case, 65% were due to indirect costs. Stratification of total costs of hospitalised cases by type of virus showed that total costs of LRTI caused by RSV were slightly higher than total costs of LRTI caused by IV, PIV or other pathogens. With the exception of bronchiolitis, the average total costs per practice case declined with decreasing severity of diagnosis. The total costs for pneumonia and obstructive bronchitis were higher than the costs for bronchitis and croup. However, the comparable low total costs for bronchiolitis was caused by lower nonmedical and indirect costs in comparison to those cost of other diagnoses. Treatment of bronchiolitis, pneumonia and obstructive bronchitis cost on average more than treatment of bronchitis or croup. In the hospital cohort, total costs per LRTI case, stratified by diagnosis, were highest in patients with apnoea and bronchiolitis. Lowest total costs were found for children with croup. Overall, the average total costs declined with decreasing severity of diagnosis.

Based on the annual incidences of community-acquired LRTI (practice cases and hospitalised cases) and nosocomial LRTI, the economic burden of LRTI in

Parameter

Office-based cases (n=654) in % (n)

Hospitalised cases (n=1118) in % (n)

Nosocomial cases (n=36) in % (n)

Additional expenses Supplemental childcare Transportation Absence from work

19.1 4.4 73.4 14.3

36.5 2.7 92.8 8.3

38.9 83.3 88.9 27.8

(125) (29) (480) (93)

(408) (30) (1038) (93)

(14) (30) (32) (10)

612 Table 5 Cost per case with community-acquired LRTI in €, office based cases and hospitalised cases Stratification by

Cost per case with community-acquired LRTI in € Office-based cases

Hospitalised cases

n

n

mean (SD) Total

Direct medical

Direct Indirect non-medical

Age group 0–12 months 13–24 months 25–36 months

221 103 (106) 220 117 (156) 127 172 (227)

68 (42) 64 (31) 72 (36)

16 (37) 12 (39) 12 (40)

Type of virus RSV PIV IV Other pathogens

115 80 17 333

79 59 66 65

(41) (28) (24) (36)

14 14 12 13

Diagnosis Apnoea 2 – – Bronchiolitis 16 94 (42) 87 Pneumonia 26 205 (264) 85 Obstructive bronchitis 247 146 (179) 83 Bronchitis 203 101 (141) 48 Croup 74 83 (78) 54 Total 568 123 (161) 67

(43) (31) (41) (21) (27) (37)

163 100 223 111

(172) (115) (280) (159)

350 74 35 463

161 131 110 127

(298) (243) (249) (258)

48 99 162 133 145 70 138

(144) (254) (299) (279) (295) (129) (281)

– 21 2 (6) 77 79 (237) 376 48 (165) 411 43 (134) 107 23 (72) 43 43 (147) 1035

Disturbances of parents’ sleep, especially for mothers, and missed usual activities of the family, especially in hospitalised LRTI cases, were reported by 37–74% of the parents (Table 7).

Discussion This evaluation assessed the average cost per patient (case) with LRTI and estimated the economic impact of LRTI due to RSV, PIV and IV in Germany. Estimation of economic impact was based on incidences generated

Direct Indirect non-medical 130 (286) 139 (248) 171 (321)

(9) (114) (39) (22) (6) (39)

Impact on patient and family activities due to LRTI

Direct medical

596 2, 827 (2213) 2, 601 (2194) 96 (171) 299 2, 242 (1132) 2, 019 (1043) 84 (112) 140 2, 243 (1335) 1, 942 (1080) 129 (253)

70 27 145 33

Germany was estimated using the median value of total costs. Economic burden due to LRTI in Germany amounted to €48.4 million for patients treated in the outpatient setting (Table 6). A total of 36% of the costs are due to RSV infections and close to 11% of the costs to PIV. The economic burden of IV was not calculated due to the small number of cases in the practice cohort (n=17). Focusing on the hospitalised cases, the economic burden due to LRTI in Germany amounted to €158.0 million annually. Hospitalisations due to RSV are responsible for 42% of the costs, to PIV and IV, 8% and 5% of the costs, respectively. Nosocomial cases incurred expenditures of €6.5 million per year. Due to the small number of nosocomial cases, economic burden was not stratified by type of virus. In total, annual expenditures due to LRTI in Germany amounted to €212.9 million.

(158) (95) (266) (145)

Total

(34) (43) (10) (41)

– 6 40 16 10 5 13

19 (78) 41 (141) 88 (221)

mean (SD)

2, 2, 2, 2,

772 374 597 482

3553 3551 2579 2609 2204 1365 2579

(1603) (1644) (1214) (2131)

(3875) (2044) (1399) (2178) (1209) (908) (1874)

2, 2, 2, 2,

507 130 428 267

3439 3170 2306 2379 1971 1255 2344

(1548) 105 (125) (1482) 113 (312) (1200) 58 (75) (2114) 89 (120)

(3873) (2017) (1303) (2150) (1079) (912) (1825)

66 81 111 97 88 39 97

(83) (89) (180) (195) (118) (40) (171)

within the population-based PRI.DE study. To our knowledge, the presented results are the first data describing the economic burden of LRTI in Germany based on epidemiological figures. Besides the economic burden, the impact on patient and family activities due to LRTI was also shown. On average, total costs incurred by office-based LRTI cases was €123 per case. More than half of the cost was reimbursed by TPP (German statutory sickness funds). The second largest proportion of costs was attributable to indirect costs due to workdays lost. However, it should be considered that the use of standardized CRFs to document resource utilization prospectively could have resulted in an underestimation of treatment cost for the office-based cases, as only pre-defined resource items were included. Due to the severity of symptoms, treatment of bronchiolitis, pneumonia and obstructive bronchitis required more health care resources from third party payers’ perspective than treatment of bronchitis and croup. Stratified by type of virus, the results show comparable costs, from the TPP’s perspective, for LRTI caused by IV and PIV, as well as for LRTI caused by other pathogens (€59–66). Treatment costs for RSVpositive infections were slightly higher (€79). Focusing on the influenza cases, 65% of the total costs per influenza case were due to indirect costs. The number of workdays lost (0.5 days) in the office-based LRTI cases is comparable to results published by Miedema and co-workers for the Netherlands [14]. In this study, parents stayed at home on average 0.5 days in the two weeks before the child’s hospital admission due to RSV infection. In the present study, total costs rose by age

a a a a

5,547,410–10,011, 705

a a a a

2913 (2146–3873)

1943

2585

7,530,105

12,056,315

9,798,549–14, 776,515

a

2487

65,965,188

59,220,444–73, 197,384

a

a

a

5,386, 671–7, 590,309 6,447,682 2100 3070 (2565–3614) 152,721,192–163, 379,736 157,993,914 2262

69,847 (67,516–72, 228) 26,524 (23,812–29, 432) 6205 (5043–7605)

Not calculated because of to small sample size

a a

IV

PIV

RSV

a

5,072,658 57

95

17,457,295

a

46,775, 794–50, 112,155 15,149, 175–19, 992,560 4,151, 367–6, 138,045 48,431,088 71

(€) (95% CI) (€)

Annual economic burden (€) Total cost median (€) (95% CI)

682,128 (658,814–705, 805) 183,761 (159,465–210, 448) 88,994 (72,831–107, 685) 25,977 (18,066–36,238) LRTI (total)

(€)

Total cost median (€) No. of cases per year (95% CI) 95% CI

Annual economic burden (€) Total cost median (€) No. of cases per year No. of cases per year

95% CI

Hospitalised cases Office-based

Table 6 Annual economic burden due to community-acquired and nosocomial LRTI in children zero to three years old

Nosocomial cases

Annual economic burden (€)

95% CI

613

group (from €103 to €172) due to increasing indirect costs. This reflects the current situation in Germany, as parents of older children are more likely to have returned to professional life than parents of infants. Total costs per hospitalised community-acquired LRTI case (€2579) and nosocomial LRTI (€2814) were dominated by the direct medical costs, which accounted for over 90%. Mean length of stay or prolongation due to LRTI was seven days. In hospitalised cases, costs were higher in children under one year of age than in older children. Treatment of apnoea and bronchiolitis required more resources than treatment of pneumonia, obstructive bronchitis, bronchitis or croup. Stratification by type of virus shows that total costs of LRTI caused by RSV (€2772) were slightly higher than total costs of LRTI caused by the other pathogens (€2374–2597). Number of workdays lost due to hospitalisation of the child (1.5 days) is similar to the results published by Miedema and coworkers [14]. Parent’s expenses are higher in both hospital cohorts than in the practice cohort. Supplemental childcare and travel expenses resulted in an additional expense of €13 for the outpatient setting and €97–113 for the hospital setting. Data on the impact on the family activities support the evaluation of the burden of disease from the parent perspective. Combined direct non-medical and indirect costs amount to €235 per child treated as inpatients, which is comparable to the results of a Dutch study, where extra expenses of €243 per child for direct non-medical and indirect cost were reported [14]. Retrospective documentation of parental expenses could introduce a potential recall bias. However, the parental expenses of hospitalised cases and nosocomial cases contributes only a small percentage to the total costs. Therefore, for this group the influence of a potential recall bias on the total cost can be neglected. To our knowledge, the present study is the first study measuring parental expenses due to LRTI for Germany. The data are a starting point for further research. The use of the human capital approach is the traditional method of calculating indirect costs in the form of overall productivity losses. This is questionable, since it discriminates against unemployed persons, pensioners and housewives as it assumes they have no market value. In particular, children in Germany up the age of three years—the study cohort—are cared for mainly by their mothers. Results of this study show hospitalisation cost of €2507 for the treatment of an LRTI caused by RSV. An average hospitalisation cost of €3615 was previously published for RSV-related hospitalisation in premature infants in Germany [18]. The cost difference can be explained with the generally higher resource use associated with prematurely born infants. In general, the comparison of treatment costs between different countries should be assessed carefully due to the different health care systems. The results of an international study show that the management of infants

614 Table 7 Impact on family life due to LRTI reported by parents in telephone interviews Parameter

Office-based cases n=654 % (n)

Hospitalised cases n=1118 % (n)

Missed day care/foster family Missed appointments of the child Missed usual activities of the family Disturbance of mother’s sleep Disturbance of father’s sleep

17.4 (113)

10.2 (113)

20.6 (134)

21.4 (236)

37.1 (240)

70.7 (778)

74.0 (477)

74.1 (819)

47.6 (302)

37.8 (405)

hospitalised with RSV infection varies markedly by country [2]. The median duration of hospitalisation in Finland and UK (four days) was lower than in Belgium, France, Germany, Italy, and the Netherlands (eight to nine days). The findings for Germany were confirmed by a German study about RSV related hospitalisations in children up to the age of 16 years, who stayed on average 9 days in hospital [29]. The median length of stay due to RSV in the presented study was about seven days. Presented costs for Germany are higher than costs in a British cohort study of children under two years of age. Direct medical costs for 653 RSV-related hospitalisations from the perspective of health authorities in UK was €834, 993 (£542,203) over a three year period, resulting in €1 279 per case [4]. The hospitalisation cost due to RSV fits more into the range of direct medical costs published for the Netherlands by Miedema and coworkers (€1572) and Rietveld and coworkers (€3110) [14, 17]. Also to be considered is the influence of certain factors on the duration of hospitalisation, as showed by a German cross-sectional study. Duration of hospitalisation was best predicted by the following parameters: young age, presence of an underlying condition, disease entity (pneumonia or bronchiolitis), intercostal retractions, and prematurity, as well as whether hospitalisation had occurred in a previous year [27]. Results of this study correspond to these findings. The youngest children (0–12 months) and children hospitalised due to more severe diagnoses like apnoea or bronchiolitis had the highest hospitalisation costs. The economic burden due to LRTI amounted to €212.9 million for patients treated in office-based practices and as inpatients (practice cases: €48.4 million; hospitalised cases: €158.0 million; nosocomial cases: €6.5 million). The population-based design of the PRI.DE study allows the study results to be generalized. Our estimate of the economic burden of LRTI is based on a comparable weak season for influenza, consequently reflecting the lower limit of the economic burden produced by health care expenditures of influenza per year in Germany [6]. Community-acquired and nosocomial LRTI in children up to age of three years cause a considerable economic burden to the health care system in Germany

annually. Preventive strategies to reduce the economic burden of LRTI are desirable. The annual number of LRTI cases in both outpatient and inpatient settings clearly demonstrates the clinical relevance of LRTI in Germany. During the two-year study period, five children died due to LRTI (four children with nosocomial infections), which underlines the clinical relevance [6]. The overall goal of health economic research is to present a set of information to clinical and political decision-makers for more rational resource allocation and optimal quality of care. This cost-of-illness study provides basic data for further decision-making, including assessment of treatment strategies for LRTI as well as vaccination programs for RSV, PIV and IV, covering inpatient and outpatient care of children up to the age of three years in Germany. Acknowledgements PRI.DE study group: Hospitals: R Berner, M Brandis, S Hameister, I Kambeck, M Posselt, S Ruess, (Freiburg); J Bittscheidt, CHL Rieger, U Schauer, J Schwarze, V Stephan (Bochum); H Gurth, K Nemat, W Leupold (Dresden); N Alex, HD Frank, V Legner (Hamburg); Paediatric practices: W Jonitz, I Keefer-Hildebrand, A Neufang (Freiburg); D Hirsch, C Hoffman, U Staemmler, E Weber (Dresden); K Heemann, J Meyer, C Simon (Bochum); H Schloon, H Sontheimer (Hamburg); Virology: H Werchau (Ruhr-Universita¨t, Bochum); Data Management: B Arnold (Freiburg); Project Management: S Merschko¨tter (Wyeth Mu¨nster). The PRI.DE study was financially supported by Wyeth GmbH, Mu¨nster, Germany. The Center for Clinical Trials receives funding by the BMBF (Federal Ministry of Education and Research).

References 1. Anderson LJ, Hendry RM, Pierik LT, Tsou C, McIntosh K (1991) Multicenter study of strains of respiratory syncytial virus. J Infect Dis 163:687–692 2. Behrendt CE, Decker MD, Burch DJ, Watson PH (1998) International variation in the management of infants hospitalized with respiratory syncytial virus. International RSV Study Group. Eur J Pediatr 157:215–220 3. Cunningham CK, McMillan JA, Gross SJ (1991) Rehospitalization for respiratory illness in infants of less than 32 weeks’ gestation. Pediatrics 88:527–532 4. Deshpande SA, Northern V (2003) The clinical and health economic burden of respiratory syncytial virus disease among children under 2 years of age in a defined geographical area. Arch Dis Child 88:1065–1069 5. Einheitlicher Bewertungsmaßstab (EBM) (2001) Ko¨ln: Deutscher A¨rzte 6. Forster J, Ihorst G, Rieger CH, Stephan V, Frank HD, Gurth H, Berner R, Rohwedder A, Werchau H, Schumacher M, Tsai T, Petersen G (2004) Prospective population-based study of viral lower respiratory tract infections in children under three years (the PRI.DE study). Eur J Pediatr 163:709–716 7. Glezen WP, Taber LH, Frank AL, Kasel JA (1986) Risk of primary infection and reinfection with respiratory syncytial virus. Am J Dis Child 140:543–546 8. Greenough A, Cox S, Alexander J, Lenney W, Turnbull F, Burgess S, Chetcuti PA, Shaw NJ, Woods A, Boorman J, Coles S, Turner J (2001) Health care utilization of infants with chronic lung disease, related to hospitalization for RSV infection. Arch Dis Child 85:463–468 9. Groothuis JR, Gutierrez KM, Lauer BA (1988) Respiratory syncytial virus infection in children with bronchopulmonary dysplasia. Pediatrics 82:199–203

615 10. Hall CB (2000) Respiratory syncytial virus. In: Mandell GL, Dolln R (eds) Principles and practice of infectious diseases. Churchill Livingstone, New York, pp 1782–1801 11. Hannoveraner Konsens Gruppe (1999) Dt. Empfehlungen zur gesundheits-o¨konomischen Evaluation—Revidierte Fassung des Hannoveraner Konsens. Gesundheitso¨konomie & Qualita¨tsmanagement 4:A62–A65 12. Ihorst G, Forster J, Petersen G, Schumacher M (2004) Determination of the population size in an epidemiological study with children. Methods Inf Med 5:479–482 13. Marchetti A, Lau H, Magar R, Wang L, Devercelli G (1999) Impact of palivizumab on expected costs of respiratory syncytial virus infection in preterm infants: potential for savings. Clin Ther 21:752–766 14. Miedema CJ, Kors AW, Tjon A, Ten WE, Kimpen JL (2001) Medical consumption and socioeconomic effects of infection with respiratory syncytial virus in The Netherlands. Pediatr Infect Dis J 20:160–163 15. Neuzil KM, Mellen BG, Wright PF, Mitchel EF, Griffin MR (2000) The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Eng J Med 342:225–231 16. Reed G, Jewett PH, Thompson J, Tollefson S, Wright PF (1997) Epidemiology and clinical impact of parainfluenza virus infections in otherwise healthy infants and young children <5 years old. JID 175:807–813 17. Rietveld E, De Jonge HC, Polder JJ, Vergouwe Y, Veeze HJ, Moll HA, Steyerberg EW (2004) Anticipated costs of hospitalization for repiratory syncytial virus infection in young children at risk. Pediatr Infect Dis J 23:523–529 18. Roeckl-Wiedmann I, Liese JG, Grill E, Fischer B, Carr D, Beloradsky BH (2003) Economic evaluation of possible prevention of RSV-related hospitalizations in premature infants in Germany. Eur J Pediatr 162:237–244 19. Rote Liste (2002) Arzneimittelverzeichnis des BPI und VFA. Bundesverband der Pharmazeutischen Industrie e.V.. Frankfurt, Aulendorf: Editio Cantor

20. Scuffham PA, West PA (2002) Economic evaluation of strategies for the control and management of influenza in Europe. Vaccine 20:2562–2578 21. Statistisches Bundesamt (2001) Statistisches Jahrbuch 2001 fu¨r die Bundesrepublik Deutschland. Wiesbaden: Metzler-Poeschel 22. Statistisches Bundesamt (2002) Statistisches Jahrbuch 2002 fu¨r die Bundesrepublik Deutschland. Wiesbaden: Metzler-Poeschel 23. Stevens TP, Sinkin RA, Hall CB, Maniscalco WM, McConnochie KM (2000) Respiratory syncytial virus and premature infants born at 32 weeks’ gestation or earlier: hospitalization and economic implications of prophylaxis. Arch Pediatr Adolesc Med 154:55–61 24. Szucs T, Behrens M, Volmer T (2001) [Public health costs of influenza in Germany 1996- a cost-of-illness analysis]. Med Klin (Munich) 96:63–70 25. The IMPACT-RSV Study Group (1998) Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants. Pediatrics 102:531–537 26. The PREVENT Study Group (1997) Reduction of respiratory syncytial virus hospitalization among premature infants and infants with bronchopulmonary dysplasia using respiratory syncytial virus immune globulin prophylaxis. Pediatrics 99:93– 99 27. Weigl JA, Puppe W, Schmitt HJ (2004) Variables explaining the duration of hospitalization in children under two years of age admitted with acute airway infections: does respiratory syncytial virus have a direct impact? Klin Pa¨diatr 216:7–15 28. Weigl JA, Puppe W, Schmitt HJ (2002) The incidence of influenza-associated hospitalizations in children in Germany. Epidemiol Infect 129:525–533 29. Weigl JA, Puppe W, Rockahr S, Schmitt HJ (2002) Burden of disease in hospitalized RSV-positive children in Germany. Klin Padiatr 214:334–342 30. Zahlenbericht der Privaten Krankenversicherer (2001/2002), Ko¨ln. www.pkv.de